- Email: [email protected]
Predicting the outcome of shunt surgery in normal pressure hydrocephalus
Journal of Clinical Neuroscience 14 (2007) 729–736 www.elsevier.com/locate/jocn
Clinical study
Predicting the outcome of shunt surgery in normal pressure hydrocephalus K. Kilic a
a,*
, A. Czorny b, J. Auque c, Z. Berkman
a
Department of Neurosurgery, Haydarpasa Numune Education and Research Hospital, Uskudar, Istanbul, Turkey b Department of Neurosurgery, University Hospital, Besanc¸on, France c Department of Neurosurgery, University Hospital, Nancy, France Received 6 February 2006; accepted 14 March 2006
Abstract We studied retrospectively the effectiveness of the repeated lumbar CSF tap test (RTT), lumbar external CSF drainage (LED) and radioisotope cisternography (RIC) in predicting the outcome of shunt surgery, as well as the diagnostic and prognostic value of periventricular hyperintensity (PVH) and of the classic clinical triad in normal pressure hydrocephalus. Two hundred and seventy patients were referred to the Departments of Neurosurgery, in Nancy, France and in Istanbul, Turkey. The decision to perform surgery was based on the clinical presentation (all patients had at least two symptoms of the classic clinical triad), neuroimaging examinations and the results of the RTT (taps were performed on three consecutive days and at each tap a minimum of 30 to 40 cc of CSF was removed), the LED (drainage was performed for 3 days and the volume of CSF drained daily was a minimum of 150 to 250 cc) or the RIC. After all shunt procedures, postoperative assessments verified improvements in 88% of the RTT group, 91% of the LED group and 66% of the RIC group. Gait disturbance had improved in 90% at the end of the second and twelfth month follow-up. Cognitive dysfunction had improved in 79% at the second and in 77% at the twelfth month follow-up. Urinary incontinence had improved in 66% at the second and in 62% at the twelfth month follow-up. From the surgical point of view, the greatest difficulty is not to make the diagnosis, but rather to identify the appropriate patients to operate on. The decision to perform shunt surgery should be based on strict clinical findings associated with CT and MRI criteria and especially with positive RTT or LED test results. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Normal pressure hydrocephalus; Adult chronic hydrocephalus; CSF tap test; Lumbar external drainage; Shunt surgery; Periventricular hyperintensity; Diagnosis
1. Introduction Normal pressure hydrocephalus (NPH) was first described by Adams and Hakim1 as a clinical syndrome with the classic symptom triad of dementia, gait disturbance, and urinary incontinence combined with widening of the cerebral ventricles, but without overt symptoms or signs of raised intracranial pressure. The symptoms are potentially reversible, and diversion of cerebrospinal fluid * Corresponding author. Address: Ziverbey 2. Hatboyu Sokak, Kilic Ap. 5/6 34724 Kiziltoprak, Istanbul, Turkey. Tel.: +90 542 425 53 86; fax: + 90 216 449 22 68. E-mail address: [email protected] (K. Kilic).
0967-5868/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jocn.2006.03.028
(CSF) either by a ventriculoperitoneal, ventriculoatrial or lumboperitoneal shunt can be successful, provided a correct diagnosis has been made. Patient selection for cerebrospinal fluid diversion in NPH is difficult and remains a diagnostic and therapeutic challenge: many patients do not display the classic clinical and neuroimaging patterns, NPH mimics other neurodegenerative disorders such as Binswanger disease (subcortical white matter arteriosclerotic encephalopathy) or Alzheimer’s disease and many patients with suspected NPH may suffer from both disorders and shunt procedures do not always result in clinical improvement. Hence, the usefulness of a shunt may be questioned. Even though the surgical procedure is technically simple, there are risks
730
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
of complications, both short- and long-term, hence various methods and clinical tests have been used to predict the outcome of a shunt procedure. In the study of any dementia or gait impairment syndrome, NPH is of particular interest as it may be treated and cured. However, prediction of which patients will benefit from surgery is one of the most important aspects of diagnosis.2 Our departments have used three methods for several years to select patients assumed to have NPH for shunt surgery: the repeated lumbar CSF tap test (RTT), lumbar external CSF drainage (LED) and radioisotope cisternography (RIC), explained in detail later. We studied retrospectively the effectiveness of RTT, LED and RIC in predicting the outcome of shunt operations, as well as the diagnostic and prognostic value of periventricular hyperintensity (PVH) and of the clinical triad in 270 unselected patients with suspected NPH. 2. Material and methods The study covers the period between December 1983 to December 2003, when 291 patients with suspected NPH were referred to the Departments of Neurosurgery at the University Hospital in Nancy, France and Haydarpasa Numune Education and Research Hospital in Istanbul, Turkey. The patients and close relatives were informed and gave their consent for all aspects of the treatment. They all fulfilled the clinical criteria of presumed NPH, but 21 had to be excluded for the following reasons: refusal to attend, intercurrent disease, changed address, missed appointments or lost data. Consequently 270 patients remained in the study and all the results refer to these patients. All patients suffered from either gait disturbance and cognitive dysfunction combined, or from one or both of them combined with urinary incontinence. Disturbance of gait was the most common preoperative symptom (250 patients, 93%), while fewer patients suffered from cognitive dysfunction (220 patients, 81%) or urinary impairment (186 patients, 69%). In every patient, different combinations of symptoms were found: 116 patients (43%) had all three symptoms of the classical triad, 154 (57%) two of them, and none (0%) only a single symptom. Eighty-four (31%) patients presented with gait abnormalities and cognitive deficits without sphincter dysfunction; 50 (19%) patients presented with gait abnormalities and sphincter dysfunction without cognitive deficits and 20 (7%) patients presented with sphincter dysfunction and cognitive deficits without gait abnormalities. CT scan or MRI was performed in all patients, and showed widening of the ventricular system with an Evan’s index >0.30 (ratio of the maximum width of the frontal horns to the maximum width of the inner table of the skull at the same level). The CT scan and MRI criteria consisted also of PVH and a more general visual evaluation, excluding patients with cortical atrophy. Frontal and occipital PVH on T2 images or hypodensities on CT scan were noted as present or absent, presence suggesting NPH. Cere-
bral atrophy was assessed by grading the cortical sulci as ‘present’ or ‘markedly enlarged’ which was interpreted as severe cortical atrophy. Cerebrovascular changes at MRI were deliberately not chosen as exclusion criteria, as neurodegenerative pathology such as Alzheimer’s disease or Binswanger disease can coexist with NPH in this age group. Therefore the ‘clinical’ diagnosis of NPH was based on the presence of the following: (i) 2–3 of the three symptoms of the triad; and (ii) enlarged ventricles at CT scan or MRI with an Evan’s index >0.30. The study included 127 men and 143 women. The mean age of the patients was 68 years (range, 48–94 years). In the majority of cases (221 patients, 82%) there was no evidence of earlier neurological disease to explain the development of NPH; they were considered as idiopathic NPH. Of 49 patients with secondary NPH (18%), 26 patients had an earlier history of head trauma, 15 patients a history of intracranial haemorrhage and eight had a history of central nervous system infection. The gait was typically slow as if the feet were stuck to the floor, unsteady, with difficulties in initiating walking (magnetic response). Gait impairment in the forms of gait imbalance (1 point), inability to walk without a cane or the aid of another person or prominent signs of dyspraxic gait instability (2 points) or total incapacity for standing or walking even with help (3 points), were all indicative of NPH and constituted the three-point gait score for all patients. In 45 patients (17%) the ‘walking test’ was used additionally as follows: the patient was asked to walk a distance of 18m as fast as possible. The number of steps as well as the time required to walk this distance before and after CSF drainage was recorded.3 Bladder function was registered as normal (0 point), increased bladder urgency (1 point), or urinary incontinence (2 points). Patients with NPH exhibited subcortical-type mental deficits including forgetfulness, decreased attention, inertia, and mental slowness with a pattern of memory impairment in delayed recall, but with preservation of delayed recognition that differed from that of Alzheimer’s disease and other cortical encephalopathies which present with aphasia, apraxia or agnosia.4,5 Walking, urinary and memory performance assessment and CT/MRI were followed by RTT, LED or RIC. The clinical results, obtained after CSF removal by RTT or LED, were compared with the best results from baseline tests. At least one-point improvement on the three-point gait score or at least 5% improvement in the walking test or at least one-point improvement on the three-point bladder function or arbitrarily an improvement of at least 30% in global mental function, alertness, responsiveness, spatialization or memory performance3 were required for the RTT or LED to be classed as positive.6 Radioisotope cisternography was carried out by injecting 99mTc-diethylene-triamine-penta-acetic acid (DTPA) via the lumbar subarachnoid space and using conventional
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
planar imaging. Between 1983 and 1993, when the RIC showed that the DTPA was detected in the ventricles and that the CSF flow over the convexity was blocked, patients were regarded as positive and they were shunted without further tests. After 1993, positive RIC alone was not considered sufficient for shunting; in addition a positive RTT or LED was required. The RTT consisted of the removal of 30–40 mL of CSF daily by lumbar puncture for three consecutive days. We used the RTT instead of a standard single CSF tap test.7,8 Lumbar external CSF drainage was performed with the patient in the lateral recumbent position. A catheter was inserted through a 16-gauge needle percutaneously in the lumbar intrathecal sac (L3–4 or L4–5). The LED system remained in situ for 3 days, if an inadvertent disconnection of the system occurred, a new system was placed. Each day, at least 150–250 mL of CSF was removed. Daily laboratory and microbiological controls of CSF samples were carried out to detect preclinical evidence of intrathecal bacterial contamination. The drainage period was arbitrarily restricted to 3 days because of the risk of test related infectious complications. The decision to perform surgery was based on the clinical presentation, neuroimaging examinations and the results of the tests. When the RTT or the LED resulted in clinical improvement of gait, urinary or mental disturbances, these tests were regarded as positive and the patients were shunted without further tests. Before 1993, positive RIC alone, and after this date, in addition to positive RIC, a positive RTT or LED was required for shunting. A total of 224 patients out of 270 who fulfilled the above criteria were operated upon. Shunts were installed right frontally in 91 patients and right occipitoparietally in 133 patients. A medium-pressure valve was used in 215 operations and low-pressure valve in only nine because of the risk of overdrainage.9 Furthermore, in two revisions, two medium-pressure valves were replaced with two low-pressure valves. Shunts were ventriculoperitoneal in 169 cases and ventriculoatrial in 55. Clinical neurological examination was carried out at 2 and 12 months after ventriculoperitoneal shunt (VPS) or ventriculoatrial shunt (VAS) surgery. Thereafter, clinical examinations, and control CT scan or MRI were carried out when shunt failure was presumed. Improvement had to be confirmed by the patients themselves, their family, or the nursing team and the patient had to improve at least one point on the gait score or at least 5% in the walking test, 30% in global mental function or at least one point on the three-point urinary functions score. The mean follow-up period was 25 months with a range of 15–42 months.
731
shunted. Following surgery 111 of these were considered as having improved when evaluated according to the criteria mentioned previously. During follow-up 10 shunts needed revision because of temporary clinical deterioration; two ventricular and eight peritoneal catheters were revised. For 18 patients who did not improve after surgery, revision of their shunt was proposed, 10 accepted, the ventricular catheter was changed in three and peritoneal in seven. Three became well after revision and seven were unchanged clinically. After shunt revisions, finally 114 of 129 patients with positive RTT were improved at the 2-month follow-up (88%). Twenty shunt revisions were performed (16%), 13 of them improved and seven did not. Twenty-six of 155 patients experienced no or questionable improvement after RTT. Even though theoretically, they should not be operated upon, a second chance of being tested by LED was systematically offered to this group of patients; 17 patients accepted this second test, seven of these improved after LED and 10 did not. Of those seven who improved after LED and were then shunted, five became better definitively, two did not benefit even after the revision of their shunt. The data from these 17 patients, who were examined by RTT and then by LED, are included and discussed in the LED group. Nine patients who did not accept LED after the negative RTT were not operated upon and were excluded from the study. 3.2. Lumbar external CSF drainage
3. Results
Seventy-three patients were studied by LED as the test of first choice. In a further 17 patients who showed either questionable or no improvement after the RTT, LED was later performed. Fifty of the 73 patients and seven of 17 improved after LED and these 57 patients were shunted. Fifty-one of these improved after the surgery according to the criteria aforementioned. During their follow-up, seven shunts needed revision because of temporary clinical deterioration, three ventricular and four peritoneal catheters were revised. For six patients who did not improve after surgery, revision of their shunt was proposed; five accepted: the ventricular catheter was changed in one and peritoneal in four. One was better after revision and four remained unchanged. After shunt revisions, 52 of 57 patients with positive LED improved clinically at the 2-month follow-up (91% of cases). Twelve shunt revisions were performed in this subgroup of 57 patients (21%), eight of them improved and four did not. Thirty-three patients with negative RTT and negative LED were not operated upon and were excluded from the follow-up.
3.1. Repeated lumbar CSF tap test
3.3. Radioisotope cisternography
This test was performed on 155 patients and resulted in clinical improvements in 129, who were consequently
Forty-two patients were studied by RIC. Ventricular filling and blockage of the CSF flow on the convexity for 48
732
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
hours were present in 38 patients (95% of 42 patients). They were all operated on and 23 of these with positive RIC improved after shunt surgery. During their followup seven shunts needed revision because of temporary clinical deterioration, two ventricular and five peritoneal catheters were revised. To the other 15 patients who showed no improvement after surgery, revision of their shunt was proposed; 10 accepted. Two were better after revision and eight remained unchanged. After shunt revisions, 25 of 38 patients with positive RIC showed improvement at the 2-months follow-up (66%). Seventeen shunt revisions were performed in this subgroup (45%); nine improved and eight did not. Four patients whose RIC was within normal limits were not operated upon and were excluded from the study. The results of surgery on 224 patients operated upon are as follows. Gait disturbance was present in 208 patients; in 187 of these (90%) this symptom had improved at the end of 2 months and this improvement was maintained at the 12-month follow-up. Cognitive dysfunction was noted in 181 patients; in 143 (79%) this had improved at 2 months and in 140 (77%) at the 12-month follow-up. Urinary incontinence was present in 155 patients; at 2 months in 102 of these (66%) it had improved and at the 12-month follow-up in 96 (62%). Of the three classic clinical symptoms, gait disturbance was most improved, both at early and late testing. Since the fibers of the corticospinal tract that supply motor function to the legs pass closest to the lateral ventricles in the corona radiata, it is not surprising that the gait disturbance is usually the first symptom to appear and the first one to be resolved following successful shunting.3 3.4. Periventricular hyperintensity This sign was present in 128 and absent in 96 of 224 patients operated upon. Seventy-one of 128 patients with PVH (55%), and 44 of 96 patients without PVH (46%) improved following surgery. This difference is not statistically significant. In agreement with our results it was demonstrated that the absence of periventricular hypodensity on CT or changes on MRI did not rule out the possibility of improvement following shunt insertion.10 3.5. Complications All patients operated upon who did not show objective improvement at the follow-up evaluation or worsening of clinical conditions after a temporary improvement had plain skull, cervical, thoracic and abdominal X-rays and CT scans of the head. In these patients shunt failure was considered due to either cranial or abdominal obstruction, or a too high opening pressure of the pump valve. In 49 out of 224 operated cases (22%) shunt placement, continuity, or function were not adequate; after revision 30 patients improved with marked benefit (61%). In 47 of these there was an obstruction or a disconnection and the medium
pressure valve was kept in place; in only two cases was it replaced with a low pressure valve and both of them benefited from this modification. In 18 out of 90 patients examined by LED (20%), the lumbar drain had to be replaced due to its disruption or displacement. In eight patients the shunting procedure was complicated by a subdural haematoma and in four by a hygroma (12/224, 5%). Four haematomas and one hygroma were managed surgically; four haematomas and three hygromas absorbed spontaneously. In four RTT (4/155, 3%), 11 LED (11/90, 12%), and 21 surgery (21/224, 9%) patients, there was related bacterial meningitis; all patients except one recovered after antibiotic therapy. The combined rate of permanent neurological deficit and death was 3% (7/224): four deaths were due to intracerebral haematoma, pulmonary embolism, nonhemorrhagic stroke and sepsis secondary to bacterial meningitis and three hemiplegias were secondary to nonhemorrhagic strokes. 4. Discussion Normal pressure hydrocephalus is in many ways a misnomer. Cerebrospinal fluid (CSF) pressure is not normal in these patients.4,11 The historical labelling NPH was based on the finding that all three reported patients by Adams and Hakim1 showed low CSF pressures at lumbar puncture, namely 150, 180, and 160 mm H2O. It is now widely recognized that a single, limited-in-time, CSF pressure measurement by lumbar puncture yields a poor estimation of the real intracranial pressure (ICP) profile of patients with NPH. There is an abundance of literature based on ICP monitoring that acknowledges that CSF pressure may not be normal in this population. From this point of view, the presence of numerous B waves on prolonged CSF pressure recording indicates a general trend to episodic high ICP.4 The general assumptions on the underlying pathophysiological mechanism include: (i) altered hydrodynamics of the CSF system, especially an insufficiency in the capacity to absorb CSF; and (ii) a parenchymal, possibly ischemic, process due to impairment of the periventricular blood flow. However, the exact mechanism of the development of the clinical symptoms is not well known.3,5 Brain perfusion studies have revealed decreased regional cerebral blood flow (CBF) in the periventricular region in patients with NPH. The decreased blood flow subsequently improves after shunt surgery. The fact that patients with idiopathic NPH respond less frequently and more transiently to shunting than patients with known causes of communicating hydrocephalus, may be because small vessel arteriosclerosis is a steadily progressive disease. Patients with the most severe white matter disease, or those with the lowest CBF, probably do not respond to shunting because irreversible atrophy has already occurred.12,13
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
Despite emerging knowledge of over 40 years and an increase in the prevalence of NPH due to longer life expectancy, its diagnosis and treatment remain controversial and the postoperative results after shunt implantation have not improved significantly. Several studies have reported that the percentage of patients who show clinical improvement after shunting is low (10–61%) and that the complication rate is excessively high (38%) with death or severe residual morbidity in 6–8% of shunted patients.5,14 For all these reasons, predictors are needed identified in order to preoperatively predict the course of the disease. Nevertheless, even today there is no agreement on the best diagnostic criteria for selecting patients with NPH for CSF shunting. In a recent study, standardised questionnaires were sent to 82 neurosurgical centres known to participate in the care of patients with NPH, the result is striking: diagnostic CSF removal was used with varying frequency by all centres except one, but the amount of CSF removed by lumbar puncture differed markedly between them; also wide variability still exists among different neurosurgical centres on the best diagnostic test for selecting patients for shunt surgery.15 Diagnostic tools for patients with suspected NPH are numerous (Table 1).3–8,10,12,14–38 Some of these require specialized equipment and their efficacy and reliability is not always established. The literature is rich in criticism for infusion tests, specific neuropsychological tests, ICP measurements, MRI investigations and for RIC. The saline infusion test with pressure monitoring has been used to reveal decreased CSF resorptive capacity.17 Data from many studies have revealed that the infusion test as well as specific neuropsychological tests are of little or no value in diagnosing NPH. Patients who should be shunted could not be distinguished from those patients who did not need a shunt by any kind of neuropsychological test.35 Meier et al. found that using the intrathecal infusion test, patients with early stage NPH, showed an improvement after a shunt operation in 65% of cases, and those with advanced stage NPH, in only 50% of cases.39 He confirmed later that outflow resistance as measured by the intrathecal infusion test showed only minimal relevance for outcome.40
Table 1 Proposed tests for diagnosing normal pressure hydrocephalus and predicting the outcome of shunt surgery Specific psychometric tests35 Periventricular hyperintensity on MRI, lucency on CT scan10,12,25,27,36–38 Measurement of CSF flow on MRI22,28,32,33 Radioisotope cisternography3,12,16,23,38 CT cisternography12,15,18,34 Single proton emission CT (SPECT)30 Measurement of regional cerebral blood flow18,31 Intrathecal infusion test with ICP measurements3,10,17,26 Ventricular tap test29 Lumbar CSF tap test3,7,8,20,23,25,34 Lumbar external CSF drainage4–6,14,19,21,24,25,38
733
For some, intracranial pressure monitoring is a useful test and B waves occurring during more than 50% of the monitoring period are associated with a likelihood of successful response to shunting.5,41 Several other authors contend that ICP measurements cannot predict outcome in NPH cases,42 that A or B-waves poorly predict which patients will respond to shunt surgery6 and that a statistically significant relationship is not found between percentage of B waves and outcome.10 It has been reported that among patients who underwent ventriculoperitoneal shunting for the treatment of NPH, measurement of CSF flow through the cerebral aqueduct on MRI did not reliably predict which patients would improve after shunting or the magnitude of improvement.5,43 It has also been stated that it is not useful to consider CSF flow void findings on conventional MRI scans in making the decision to offer shunting to patients with idiopathic NPH.28 Nuclear or CT cisternography has been proposed for the evaluation of NPH. If this test shows ventricular reflux with slow cortical uptake, it reveals disordered CSF reabsorption at the level of the arachnoid villi and may have a role in diagnosing communicating hydrocephalus, but it cannot predict shunt response. Its presence does not reliably predict which patients will have favourable outcomes after shunt surgery; similarly, a negative test does not preclude a favourable response to shunt surgery.12,16,23,25 The results of the present series also support this fact: only 66% of patients with positive RIC improved after the shunt surgery. In fact, after 1993, we did not consider a positive RIC sufficient for shunting, and therefore we required an additional positive RTT or LED. It has been published that diagnostic ventricular CSF removal may be a valuable ancillary test in the selection of patients for shunt surgery and that this ‘ventricular tap test’ has shown much greater sensitivity and specificity in selecting which patients will respond to shunting, which is not surprising given that the test comes closest to simulating the actual VP shunt.29 The RTT is an important diagnositc test because of its simplicity and less invasive nature.16,25 It is generally assumed that of the diagnostic studies, the most reliable result is improvement in clinical symptoms following a lumbar puncture in which CSF is withdrawn,20,34 and Milhorat reported that a favourable response to preoperative CSF removal is the most reliable indicator of surgical outcome in patients with NPH.44 Analysis of data from several studies has demonstrated that LED is an accurate test for predicting outcome after ventricular shunting in patients with NPH.6,19,21,24,25 The results of the present study confirm the accuracy of RTT and LED in selecting patients for shunt surgery, with an improvement rate of 88% and 91%, respectively. In the series of Krauss, with the lumbar tap test used as the diagnostic tool provided a similar improvement of 90%.27 In some studies, false negative predictions were high, with the tap test at 58%,3 and the lumbar external drainage
734
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
at 64% false negative predictions.38 It seems that these tests, in their originally described forms, are not accurate enough to reproduce the conditions of a patient operated upon with VPS. Since the 1980s we have used these simple but useful tests with modifications. First, in RTT, the numbers of days where taps are performed was increased to three consecutive days and at each tap a minimum of 30–40 cc CSF was removed. Second, in LED, the volume of CSF drained daily was a minimum of 150–250 cc and this during three consecutive days. Our series has two further differences from many recent reports. First, our patients had at least two symptoms of the clinical triad. We encounter more and more articles where patients with only gait disturbances have been operated upon: this trend of considering NPH primarily as a treatable gait disorder is as dangerous as to consider it primarily as a treatable dementia, much as it was 20 years ago. The high rate of unfavourable outcomes in these series seems to us inevitable: it is well established that post-surgical improvement is greater in patients with the complete clinical triad compared with those who had presented with incomplete forms of the clinical syndrome.10,14,16,35,45,46 The clinical triad is necessary but insufficient, for two reasons: first, it can be encountered in pathologies other than NPH, such as any form of chronic, communicating hydrocephalus, and even a few non-communicating forms such as aqueduct stenosis. All these patients may present with a similar clinical triad, and they may all be treated with a ventriculoperitoneal shunt.12 Cerebral atrophy and deep white matter ischemia are both far more common causes of the clinical triad than NPH.23 Second, the rate of clinical improvement and the positive predictive value of the complete triad are respectively not more than 61% and 65% in the series of NPH patients.16,37 Therefore, the use of a valuable complementary clinical test is mandatory to obtain better results than 65%. Finally, in all operated patients who did not show objective improvement at the follow-up evaluation or worsening of clinical conditions after a temporary improvement, shunt failure was considered and the shunt system was revised in every patient who consented to revision. Ineffective shunt function is a frequent cause of poor outcome after shunt surgery; one must confirm that the shunt is functioning before the diagnosis of NPH is labelled ‘incorrect’. Williams stated in 1998 that ‘poor clinical outcome occurred in two-thirds of all patients after shunt surgery for NPH, but a potentially treatable cause was found in nearly 80% of shunt systems investigated and clinical recovery occurred in 70% of patients who had shunt revision surgery’.47 In the present series, in 49 cases shunt revision was necessary, and after revision, 30 patients (61%) improved with marked benefit. Hebb stated in his review of 44 articles14 an overall improvement of 59% after shunting and only 29% of prolonged improvement. Our results may suggest that with
some modifications in diagnostic tests as mentioned earlier, and with more rigorous patient selection criteria, these results of shunt surgery can be improved in NPH patients. A criticism of this highly selective attitude can be that we excluded some patients who may have benefited from surgery, but the reason for this policy is again the unacceptably high rate of complications, permanent neurological deficit and death in this group of patients and our intention thereby of minimizing unnecessary operations. Otherwise the outcome could have been 6–8% combined rate of permanent neurological deficit and death as in Hebb’s and Vanneste’s studies.5,14 Alzheimer’s and Binswanger’s disease are common sources of comorbidity in older patients with NPH where they contribute to the clinical impairment associated with this disorder. A number of authors have commented on the comorbidities. ‘For patients accurately diagnosed with NPH, concomitant Alzheimer’s disease pathology does not strongly influence the clinical response to shunt surgery.’27,48 ‘Cortical atrophy does not rule out the possibility of improvement after shunt placement, patients should not be denied surgery solely because of dilated sulci or fissures.’10 Our study seems to confirm both these statements: two patients with Parkinson’s disease, four with Alzheimer’s disease and 14 with different forms of sequelae of cerebrovascular accident are of particular interest, as two, two and eight of them respectively, recovered from their NPH symptoms after shunt surgery. 5. Conclusion The results suggest that both the RTT and the LED can accurately predict the outcome of shunt surgery prior to shunt procedures in patients with suspected NPH. The low costs and the simplicity of these tests compared to the invasiveness and the possibility of serious test-related complications of other tests show their usefulness in managing patients with presumed NPH. It has been shown that good results after shunting can be expected in patients showing improvement after a CSF tap test, so surgery may be based on a positive RTT alone. However, if negative, RTT may be complemented by a 3-day LED which is another reliable predictor of favourable outcome after shunting.4,6,14,19,24,25,38 The presence of PVH or subcortical lacunar infarctions in NPH, although of negative prognostic value, does not predict a poor outcome from shunt surgery and should not be used as exclusion criteria for shunting.27,36 From a surgical point of view, the real problem is not to determine the exact diagnostic label, but to identify the appropriate patient to operate on; RTT or LED, when performed in clinically selected patients, seem to provide this information. Our results show that a high percentage of improvement with a low mortality rate and few complications can be achieved in patients with NPH using a rigorous clinical protocol associated with CT/MRI criteria and particularly with positive RTT or LED test results. Our experience
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
and that of others27 indicate that the percentage of improvement after shunting can be as high as 90%, on the condition that this strict diagnostic and treatment protocol is applied. Radioisotope cisternography, lumbar infusion tests and continuous ICP recording can all show that CSF dynamics is ‘deficient or malfunctioning’; they may even be able to establish that the diagnosis is NPH, but none can confirm directly whether the patient will benefit from surgery. In other words, RTT and LED, instead of answering the question, that is of secondary importance to both surgeon and patient, ‘does the patient have NPH or not’, rather responds directly to the question of primary importance: ‘will the patient benefit from this operation, and should they consequently, undergo surgery’. Therefore, the prediction of post-operative-like results, as if surgery has been performed, becomes an invaluable foresight, an opportunity which cannot be encountered in any other situation within the field of neurosurgery. References 1. Adams RD, Fisher CM, Hakim S, et al. Symptomatic occult hydrocephalus with ‘‘Normal’’ cerebrospinal fluid pressure: a treatable syndrome. N Engl J Med 1965;273:117–26. 2. Arriada-Mendicoa N, Herrera-Guerrero MP, Otero-Siliceo E. Chronic hydrocephaly in adults. A diagnostic and therapeutic challenge. Rev Neurol 2002;34:665–72. 3. Kahlon B, Sundba¨rg G, Rehncrona S. Comparison between the lumbar infusion and CSF tap tests to predict outcome after shunt surgery in suspected normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry 2002;73:721–6. 4. Bret P, Guyotat J, Chazal J. Is normal pressure hydrocephalus a valid concept in 2002? A reappraisal in five questions and proposal for a new designation of the syndrome as ‘chronic hydrocephalus’. J Neurol Neurosurg Psychiatry 2002;73:9–12. 5. Vanneste JA. Diagnosis and management of normal-pressure hydrocephalus. J Neurol 2000;247:5–14. 6. Williams MA, Razumovsky AY, Hanley DF. Comparison of PCSF monitoring and controlled CSF drainage diagnose normal pressure hydrocephalus. Acta Neurochir Suppl. 1998;71:328–30. 7. Takeuchi T, Iwasaki M, Shirata K. Evaluation of lumbar tap test combined with mean cerebral blood flow measurement and electroencephalographic topography on idiopathic normal pressure hydrocephalus patients. No Shinkei Geka 2004;32 (3):247–55. 8. Wikkelso¨ C, Andersson H, Blomstrand C, et al. The clinical effect of lumbar puncture in normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry 1982;45:64–9. 9. De Jong DA, Delwel EJ, Avezaat CJ. Hydrostatic and hydrodynamic considerations in shunted normal pressure hydrocephalus. Acta Neurochir (Wien) 2000;142 (3):241–7. 10. Poca MA, Mataro M, Matarin MM, et al. Is the placement of shunts in patients with idiopathic normal-pressure hydrocephalus worth the risk? Results of a study based on continuous monitoring of intracranial pressure. J Neurosurg. 2004;100:855–66. 11. Dunn L. ‘Normal pressure hydrocephalus’: what’s in a name? J Neurol Neurosurg Psychiatry 2002;73:8. 12. Bradley WG. Normal pressure hydrocephalus: new concepts on etiology and diagnosis. Am J Neuroradiol 2000;21:1586–90. 13. Bradley WG. Normal pressure hydrocephalus and deep white matter ischemia: which is the chicken, and which is the egg? Am J Neuroradiol 2001;22:1638–40. 14. Hebb AO, Cusimano MD. Idiopathic normal pressure hydrocephalus: a systematic review of diagnosis and outcome. Neurosurgery 2001;49:1166–84, discussion 1184–6.
735
15. Krauss JK, Halve B. Normal pressure hydrocephalus: survey on contemporary diagnostic algorithms and therapeutic decision-making in clinical practice. Acta Neurochir (Wien) 2004;146:379–88, discussion 388. 16. Black PM. Idiopathic normal-pressure hydrocephalus. Results of shunting in 62 patients. J Neurosurg 1980;52:371–7. 17. Boon AJ, Tans JT, Delwel EJ, et al. Does CSF outflow resistance predict the response to shunting in patients with normal pressure hydrocephalus? Acta Neurochir Suppl (Wien) 1998;71:331–3. 18. Chang CC, Kuwana N, Ito S, et al. Prediction of effectiveness of shunting in patients with normal pressure hydrocephalus by cerebral blood flow measurement and computed tomography cisternography. Neurol Med Chir (Tokyo) 1999;39:841–5, discussion, 845–6. 19. Chen IH, Huang CI, Liu HC, et al. Effectiveness of shunting in patients with normal pressure hydrocephalus predicted by temporary, controlled-resistance, continuous lumbar drainage: a pilot study. J Neurol Neurosurg Psychiatry 1994;57:1430–2. 20. Damasceno BP, Carelli EF, Honorato DC, et al. The predictive value of cerebrospinal fluid tap-test in normal pressure hydrocephalus. Arq Neuropsiquiatr. 1997;55:179–85. 21. Di Lauro L, Viearini M, Bollati A. The predictive value of 5 days CSF diversion for shunting in normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry 1986;49:842–3. 22. Egeler-Peerdeman SM, Barkhof F, Walchenbach R, et al. Cine phasecontrast MR imaging in normal pressure hydrocephalus patients: relation to surgical outcome. Acta Neurochir Suppl (Wien) 1998;71: 340–2. 23. Fishman RA, Dillon WP. Normal pressure hydrocephalus: new findings and old questions. Am J Neuroradiol 2001;22:1640–1. 24. Haan J, Thomeer RTWM. Predictive value of temporary external lumbar drainage in normal pressure hydrocephalus. Neurosurgery 1988;22:388–91. 25. Ishikawa M. Guideline Committee for Idiopathic Normal Pressure Hydrocephalus, Japanese Society of Normal Pressure Hydrocephalus. Clinical guidelines for idiopathic normal pressure hydrocephalus. Neurol Med Chir (Tokyo) 2004;44:222–3. 26. Kosteljanetz M, Nehen AM, Kaalund J. Cerebrospinal fluid outflow resistance measurements in the selection of patients for shunt surgery in the normal pressure hydrocephalus syndrome. A controlled trial. Acta Neurochir 1990;104:48–53. 27. Krauss JK, Droste DW, Vach W, et al. Cerebrospinal fluid shunting in idiopathic normal-pressure hydrocephalus of the elderly: effect of periventricular and deep white matter lesions. Neurosurgery 1996;39: 292–9, discussion 299–300. 28. Krauss JK, Regel JP, Vach W, et al. Flow void of cerebrospinal fluid in idiopathic normal pressure hydrocephalus of the elderly: can it predict outcome after shunting? Neurosurgery 1997;40:67–73, discussion 73–4. 29. Krauss JK, Regel JP. The predictive value of ventricular CSF removal in normal pressure hydrocephalus. Neurol Res 1997;19:357–60. 30. Larsson A, Arlig A, Bergh AC, et al. Quantitative SPECT cisternography in normal pressure hydrocephalus. Acta Neurol Scand 1994;90:190–6. 31. Larsson A, Bergh AC, Bilting M, et al. Regional cerebral blood flow in normal pressure hydrocephalus: diagnostic and prognostic aspects. Eur J Nucl Med 1994;21:118–23. 32. Luertmer PH, Huston J, Friedman JA, et al. Measurement of cerebrospinal fluid flow at the cerebral aqueduct by use of phasecontrast magnetic resonance imaging: technique validation and utility in diagnosing idiopathic normal pressure hydrocephalus. Neurosurgery 2002;50:534–43, discussion 543–4. 33. Mase M, Yamada K, Banno T, et al. Quantitative analysis of CSF flow dynamics using MRI in normal pressure hydrocephalus. Acta Neurochir Suppl (Wien) 1998;71:350–3. 34. Mori K. Management of idiopathic normal-pressure hydrocephalus: a multiinstitutional study conducted in Japan. J Neurosurg 2001;95: 970–3.
736
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
35. Savolainen S, Hurskainen H, Paljarvi L, et al. Five-year outcome of normal pressure hydrocephalus with or without a shunt: predictive value of the clinical signs, neuropsychological evaluation and infusion test. Acta Neurochir (Wien) 2002;144:515–23, discussion 523. 36. Tullberg M, Jensen C, Ekholm S, et al. Normal pressure hydrocephalus: Vascular white matter changes on MR images must not exclude patients from shunt surgery. Am J Neuroradiol 2001;22:1665–73. 37. Vanneste J, Augustijn P, Tan WF, et al. Shunting normal pressure hydrocephalus: the predictive value of combined clinical and CT data. J Neurol Neurosurg Psychiatry 1993;56:251–6. 38. Walchenbach R, Geiger E, Thomeer RT, et al. The value of temporary external lumbar CSF drainage in predicting the outcome of shunting on normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry 2002;72:503–6. 39. Meier U, Bartels P. The importance of the intrathecal infusion test in the diagnosis of normal pressure hydrocephalus. J Clin Neurosci 2002;9:260–7. 40. Meier U, Konig A, Miethke C. Predictors of outcome in patients with normal-pressure hydrocephalus. Eur Neurol 2004;51:59–67. 41. Reilly P. In normal pressure hydrocephalus, intracranial pressure monitoring is the only useful test. J Clin Neurosci 2001;8:66–7.
42. Sorteberg A, Eide PK, Fremming AD. A prospective study on the clinical effect of surgical treatment of normal pressure hydrocephalus: the value of hydrodynamic evaluation. Br J Neurosurg 2004;18: 149–57. 43. Dixon GR, Friedman JA, Luetmer PH, et al. Use of cerebrospinal fluid flow rates measured by phase-contrast MR to predict outcome of ventriculoperitoneal shunting for idiopathic normal-pressure hydrocephalus. Mayo Clin Proc 2002;77:509–14. 44. Milhorat TH. Comment. Neurosurgery 1997;40:73–4. 45. Janny P, Colnet P, Veyra A, et al. Hydroce´phalie a` pression normale. Etude pre´ et postope´ratoire de 56 cas. Neurochirurgie 1981;27:89–96. 46. Larsson A, Wikkelso¨ C, Bilting M, et al. Clinical parameters in 74 consecutive patients shunt operated for normal pressure hydrocephalus. Acta Neurol Scand 1991;84:475–82. 47. Williams MA, Razumovsky AY, Hanley DF. Evaluation of shunt function in patients who are never better, or better then worse after shunt surgery for NPH. Acta Neurochir Suppl (Wien) 1998;71: 368–70. 48. Golomb J, Wisoff J, Miller DC, et al. Alzheimer’s disease comorbidity in normal pressure hydrocephalus: prevalence and shunt response. J Neurol Neurosurg Psychiatry 2000;68:778–81.
Clinical study
Predicting the outcome of shunt surgery in normal pressure hydrocephalus K. Kilic a
a,*
, A. Czorny b, J. Auque c, Z. Berkman
a
Department of Neurosurgery, Haydarpasa Numune Education and Research Hospital, Uskudar, Istanbul, Turkey b Department of Neurosurgery, University Hospital, Besanc¸on, France c Department of Neurosurgery, University Hospital, Nancy, France Received 6 February 2006; accepted 14 March 2006
Abstract We studied retrospectively the effectiveness of the repeated lumbar CSF tap test (RTT), lumbar external CSF drainage (LED) and radioisotope cisternography (RIC) in predicting the outcome of shunt surgery, as well as the diagnostic and prognostic value of periventricular hyperintensity (PVH) and of the classic clinical triad in normal pressure hydrocephalus. Two hundred and seventy patients were referred to the Departments of Neurosurgery, in Nancy, France and in Istanbul, Turkey. The decision to perform surgery was based on the clinical presentation (all patients had at least two symptoms of the classic clinical triad), neuroimaging examinations and the results of the RTT (taps were performed on three consecutive days and at each tap a minimum of 30 to 40 cc of CSF was removed), the LED (drainage was performed for 3 days and the volume of CSF drained daily was a minimum of 150 to 250 cc) or the RIC. After all shunt procedures, postoperative assessments verified improvements in 88% of the RTT group, 91% of the LED group and 66% of the RIC group. Gait disturbance had improved in 90% at the end of the second and twelfth month follow-up. Cognitive dysfunction had improved in 79% at the second and in 77% at the twelfth month follow-up. Urinary incontinence had improved in 66% at the second and in 62% at the twelfth month follow-up. From the surgical point of view, the greatest difficulty is not to make the diagnosis, but rather to identify the appropriate patients to operate on. The decision to perform shunt surgery should be based on strict clinical findings associated with CT and MRI criteria and especially with positive RTT or LED test results. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Normal pressure hydrocephalus; Adult chronic hydrocephalus; CSF tap test; Lumbar external drainage; Shunt surgery; Periventricular hyperintensity; Diagnosis
1. Introduction Normal pressure hydrocephalus (NPH) was first described by Adams and Hakim1 as a clinical syndrome with the classic symptom triad of dementia, gait disturbance, and urinary incontinence combined with widening of the cerebral ventricles, but without overt symptoms or signs of raised intracranial pressure. The symptoms are potentially reversible, and diversion of cerebrospinal fluid * Corresponding author. Address: Ziverbey 2. Hatboyu Sokak, Kilic Ap. 5/6 34724 Kiziltoprak, Istanbul, Turkey. Tel.: +90 542 425 53 86; fax: + 90 216 449 22 68. E-mail address: [email protected] (K. Kilic).
0967-5868/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jocn.2006.03.028
(CSF) either by a ventriculoperitoneal, ventriculoatrial or lumboperitoneal shunt can be successful, provided a correct diagnosis has been made. Patient selection for cerebrospinal fluid diversion in NPH is difficult and remains a diagnostic and therapeutic challenge: many patients do not display the classic clinical and neuroimaging patterns, NPH mimics other neurodegenerative disorders such as Binswanger disease (subcortical white matter arteriosclerotic encephalopathy) or Alzheimer’s disease and many patients with suspected NPH may suffer from both disorders and shunt procedures do not always result in clinical improvement. Hence, the usefulness of a shunt may be questioned. Even though the surgical procedure is technically simple, there are risks
730
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
of complications, both short- and long-term, hence various methods and clinical tests have been used to predict the outcome of a shunt procedure. In the study of any dementia or gait impairment syndrome, NPH is of particular interest as it may be treated and cured. However, prediction of which patients will benefit from surgery is one of the most important aspects of diagnosis.2 Our departments have used three methods for several years to select patients assumed to have NPH for shunt surgery: the repeated lumbar CSF tap test (RTT), lumbar external CSF drainage (LED) and radioisotope cisternography (RIC), explained in detail later. We studied retrospectively the effectiveness of RTT, LED and RIC in predicting the outcome of shunt operations, as well as the diagnostic and prognostic value of periventricular hyperintensity (PVH) and of the clinical triad in 270 unselected patients with suspected NPH. 2. Material and methods The study covers the period between December 1983 to December 2003, when 291 patients with suspected NPH were referred to the Departments of Neurosurgery at the University Hospital in Nancy, France and Haydarpasa Numune Education and Research Hospital in Istanbul, Turkey. The patients and close relatives were informed and gave their consent for all aspects of the treatment. They all fulfilled the clinical criteria of presumed NPH, but 21 had to be excluded for the following reasons: refusal to attend, intercurrent disease, changed address, missed appointments or lost data. Consequently 270 patients remained in the study and all the results refer to these patients. All patients suffered from either gait disturbance and cognitive dysfunction combined, or from one or both of them combined with urinary incontinence. Disturbance of gait was the most common preoperative symptom (250 patients, 93%), while fewer patients suffered from cognitive dysfunction (220 patients, 81%) or urinary impairment (186 patients, 69%). In every patient, different combinations of symptoms were found: 116 patients (43%) had all three symptoms of the classical triad, 154 (57%) two of them, and none (0%) only a single symptom. Eighty-four (31%) patients presented with gait abnormalities and cognitive deficits without sphincter dysfunction; 50 (19%) patients presented with gait abnormalities and sphincter dysfunction without cognitive deficits and 20 (7%) patients presented with sphincter dysfunction and cognitive deficits without gait abnormalities. CT scan or MRI was performed in all patients, and showed widening of the ventricular system with an Evan’s index >0.30 (ratio of the maximum width of the frontal horns to the maximum width of the inner table of the skull at the same level). The CT scan and MRI criteria consisted also of PVH and a more general visual evaluation, excluding patients with cortical atrophy. Frontal and occipital PVH on T2 images or hypodensities on CT scan were noted as present or absent, presence suggesting NPH. Cere-
bral atrophy was assessed by grading the cortical sulci as ‘present’ or ‘markedly enlarged’ which was interpreted as severe cortical atrophy. Cerebrovascular changes at MRI were deliberately not chosen as exclusion criteria, as neurodegenerative pathology such as Alzheimer’s disease or Binswanger disease can coexist with NPH in this age group. Therefore the ‘clinical’ diagnosis of NPH was based on the presence of the following: (i) 2–3 of the three symptoms of the triad; and (ii) enlarged ventricles at CT scan or MRI with an Evan’s index >0.30. The study included 127 men and 143 women. The mean age of the patients was 68 years (range, 48–94 years). In the majority of cases (221 patients, 82%) there was no evidence of earlier neurological disease to explain the development of NPH; they were considered as idiopathic NPH. Of 49 patients with secondary NPH (18%), 26 patients had an earlier history of head trauma, 15 patients a history of intracranial haemorrhage and eight had a history of central nervous system infection. The gait was typically slow as if the feet were stuck to the floor, unsteady, with difficulties in initiating walking (magnetic response). Gait impairment in the forms of gait imbalance (1 point), inability to walk without a cane or the aid of another person or prominent signs of dyspraxic gait instability (2 points) or total incapacity for standing or walking even with help (3 points), were all indicative of NPH and constituted the three-point gait score for all patients. In 45 patients (17%) the ‘walking test’ was used additionally as follows: the patient was asked to walk a distance of 18m as fast as possible. The number of steps as well as the time required to walk this distance before and after CSF drainage was recorded.3 Bladder function was registered as normal (0 point), increased bladder urgency (1 point), or urinary incontinence (2 points). Patients with NPH exhibited subcortical-type mental deficits including forgetfulness, decreased attention, inertia, and mental slowness with a pattern of memory impairment in delayed recall, but with preservation of delayed recognition that differed from that of Alzheimer’s disease and other cortical encephalopathies which present with aphasia, apraxia or agnosia.4,5 Walking, urinary and memory performance assessment and CT/MRI were followed by RTT, LED or RIC. The clinical results, obtained after CSF removal by RTT or LED, were compared with the best results from baseline tests. At least one-point improvement on the three-point gait score or at least 5% improvement in the walking test or at least one-point improvement on the three-point bladder function or arbitrarily an improvement of at least 30% in global mental function, alertness, responsiveness, spatialization or memory performance3 were required for the RTT or LED to be classed as positive.6 Radioisotope cisternography was carried out by injecting 99mTc-diethylene-triamine-penta-acetic acid (DTPA) via the lumbar subarachnoid space and using conventional
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
planar imaging. Between 1983 and 1993, when the RIC showed that the DTPA was detected in the ventricles and that the CSF flow over the convexity was blocked, patients were regarded as positive and they were shunted without further tests. After 1993, positive RIC alone was not considered sufficient for shunting; in addition a positive RTT or LED was required. The RTT consisted of the removal of 30–40 mL of CSF daily by lumbar puncture for three consecutive days. We used the RTT instead of a standard single CSF tap test.7,8 Lumbar external CSF drainage was performed with the patient in the lateral recumbent position. A catheter was inserted through a 16-gauge needle percutaneously in the lumbar intrathecal sac (L3–4 or L4–5). The LED system remained in situ for 3 days, if an inadvertent disconnection of the system occurred, a new system was placed. Each day, at least 150–250 mL of CSF was removed. Daily laboratory and microbiological controls of CSF samples were carried out to detect preclinical evidence of intrathecal bacterial contamination. The drainage period was arbitrarily restricted to 3 days because of the risk of test related infectious complications. The decision to perform surgery was based on the clinical presentation, neuroimaging examinations and the results of the tests. When the RTT or the LED resulted in clinical improvement of gait, urinary or mental disturbances, these tests were regarded as positive and the patients were shunted without further tests. Before 1993, positive RIC alone, and after this date, in addition to positive RIC, a positive RTT or LED was required for shunting. A total of 224 patients out of 270 who fulfilled the above criteria were operated upon. Shunts were installed right frontally in 91 patients and right occipitoparietally in 133 patients. A medium-pressure valve was used in 215 operations and low-pressure valve in only nine because of the risk of overdrainage.9 Furthermore, in two revisions, two medium-pressure valves were replaced with two low-pressure valves. Shunts were ventriculoperitoneal in 169 cases and ventriculoatrial in 55. Clinical neurological examination was carried out at 2 and 12 months after ventriculoperitoneal shunt (VPS) or ventriculoatrial shunt (VAS) surgery. Thereafter, clinical examinations, and control CT scan or MRI were carried out when shunt failure was presumed. Improvement had to be confirmed by the patients themselves, their family, or the nursing team and the patient had to improve at least one point on the gait score or at least 5% in the walking test, 30% in global mental function or at least one point on the three-point urinary functions score. The mean follow-up period was 25 months with a range of 15–42 months.
731
shunted. Following surgery 111 of these were considered as having improved when evaluated according to the criteria mentioned previously. During follow-up 10 shunts needed revision because of temporary clinical deterioration; two ventricular and eight peritoneal catheters were revised. For 18 patients who did not improve after surgery, revision of their shunt was proposed, 10 accepted, the ventricular catheter was changed in three and peritoneal in seven. Three became well after revision and seven were unchanged clinically. After shunt revisions, finally 114 of 129 patients with positive RTT were improved at the 2-month follow-up (88%). Twenty shunt revisions were performed (16%), 13 of them improved and seven did not. Twenty-six of 155 patients experienced no or questionable improvement after RTT. Even though theoretically, they should not be operated upon, a second chance of being tested by LED was systematically offered to this group of patients; 17 patients accepted this second test, seven of these improved after LED and 10 did not. Of those seven who improved after LED and were then shunted, five became better definitively, two did not benefit even after the revision of their shunt. The data from these 17 patients, who were examined by RTT and then by LED, are included and discussed in the LED group. Nine patients who did not accept LED after the negative RTT were not operated upon and were excluded from the study. 3.2. Lumbar external CSF drainage
3. Results
Seventy-three patients were studied by LED as the test of first choice. In a further 17 patients who showed either questionable or no improvement after the RTT, LED was later performed. Fifty of the 73 patients and seven of 17 improved after LED and these 57 patients were shunted. Fifty-one of these improved after the surgery according to the criteria aforementioned. During their follow-up, seven shunts needed revision because of temporary clinical deterioration, three ventricular and four peritoneal catheters were revised. For six patients who did not improve after surgery, revision of their shunt was proposed; five accepted: the ventricular catheter was changed in one and peritoneal in four. One was better after revision and four remained unchanged. After shunt revisions, 52 of 57 patients with positive LED improved clinically at the 2-month follow-up (91% of cases). Twelve shunt revisions were performed in this subgroup of 57 patients (21%), eight of them improved and four did not. Thirty-three patients with negative RTT and negative LED were not operated upon and were excluded from the follow-up.
3.1. Repeated lumbar CSF tap test
3.3. Radioisotope cisternography
This test was performed on 155 patients and resulted in clinical improvements in 129, who were consequently
Forty-two patients were studied by RIC. Ventricular filling and blockage of the CSF flow on the convexity for 48
732
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
hours were present in 38 patients (95% of 42 patients). They were all operated on and 23 of these with positive RIC improved after shunt surgery. During their followup seven shunts needed revision because of temporary clinical deterioration, two ventricular and five peritoneal catheters were revised. To the other 15 patients who showed no improvement after surgery, revision of their shunt was proposed; 10 accepted. Two were better after revision and eight remained unchanged. After shunt revisions, 25 of 38 patients with positive RIC showed improvement at the 2-months follow-up (66%). Seventeen shunt revisions were performed in this subgroup (45%); nine improved and eight did not. Four patients whose RIC was within normal limits were not operated upon and were excluded from the study. The results of surgery on 224 patients operated upon are as follows. Gait disturbance was present in 208 patients; in 187 of these (90%) this symptom had improved at the end of 2 months and this improvement was maintained at the 12-month follow-up. Cognitive dysfunction was noted in 181 patients; in 143 (79%) this had improved at 2 months and in 140 (77%) at the 12-month follow-up. Urinary incontinence was present in 155 patients; at 2 months in 102 of these (66%) it had improved and at the 12-month follow-up in 96 (62%). Of the three classic clinical symptoms, gait disturbance was most improved, both at early and late testing. Since the fibers of the corticospinal tract that supply motor function to the legs pass closest to the lateral ventricles in the corona radiata, it is not surprising that the gait disturbance is usually the first symptom to appear and the first one to be resolved following successful shunting.3 3.4. Periventricular hyperintensity This sign was present in 128 and absent in 96 of 224 patients operated upon. Seventy-one of 128 patients with PVH (55%), and 44 of 96 patients without PVH (46%) improved following surgery. This difference is not statistically significant. In agreement with our results it was demonstrated that the absence of periventricular hypodensity on CT or changes on MRI did not rule out the possibility of improvement following shunt insertion.10 3.5. Complications All patients operated upon who did not show objective improvement at the follow-up evaluation or worsening of clinical conditions after a temporary improvement had plain skull, cervical, thoracic and abdominal X-rays and CT scans of the head. In these patients shunt failure was considered due to either cranial or abdominal obstruction, or a too high opening pressure of the pump valve. In 49 out of 224 operated cases (22%) shunt placement, continuity, or function were not adequate; after revision 30 patients improved with marked benefit (61%). In 47 of these there was an obstruction or a disconnection and the medium
pressure valve was kept in place; in only two cases was it replaced with a low pressure valve and both of them benefited from this modification. In 18 out of 90 patients examined by LED (20%), the lumbar drain had to be replaced due to its disruption or displacement. In eight patients the shunting procedure was complicated by a subdural haematoma and in four by a hygroma (12/224, 5%). Four haematomas and one hygroma were managed surgically; four haematomas and three hygromas absorbed spontaneously. In four RTT (4/155, 3%), 11 LED (11/90, 12%), and 21 surgery (21/224, 9%) patients, there was related bacterial meningitis; all patients except one recovered after antibiotic therapy. The combined rate of permanent neurological deficit and death was 3% (7/224): four deaths were due to intracerebral haematoma, pulmonary embolism, nonhemorrhagic stroke and sepsis secondary to bacterial meningitis and three hemiplegias were secondary to nonhemorrhagic strokes. 4. Discussion Normal pressure hydrocephalus is in many ways a misnomer. Cerebrospinal fluid (CSF) pressure is not normal in these patients.4,11 The historical labelling NPH was based on the finding that all three reported patients by Adams and Hakim1 showed low CSF pressures at lumbar puncture, namely 150, 180, and 160 mm H2O. It is now widely recognized that a single, limited-in-time, CSF pressure measurement by lumbar puncture yields a poor estimation of the real intracranial pressure (ICP) profile of patients with NPH. There is an abundance of literature based on ICP monitoring that acknowledges that CSF pressure may not be normal in this population. From this point of view, the presence of numerous B waves on prolonged CSF pressure recording indicates a general trend to episodic high ICP.4 The general assumptions on the underlying pathophysiological mechanism include: (i) altered hydrodynamics of the CSF system, especially an insufficiency in the capacity to absorb CSF; and (ii) a parenchymal, possibly ischemic, process due to impairment of the periventricular blood flow. However, the exact mechanism of the development of the clinical symptoms is not well known.3,5 Brain perfusion studies have revealed decreased regional cerebral blood flow (CBF) in the periventricular region in patients with NPH. The decreased blood flow subsequently improves after shunt surgery. The fact that patients with idiopathic NPH respond less frequently and more transiently to shunting than patients with known causes of communicating hydrocephalus, may be because small vessel arteriosclerosis is a steadily progressive disease. Patients with the most severe white matter disease, or those with the lowest CBF, probably do not respond to shunting because irreversible atrophy has already occurred.12,13
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
Despite emerging knowledge of over 40 years and an increase in the prevalence of NPH due to longer life expectancy, its diagnosis and treatment remain controversial and the postoperative results after shunt implantation have not improved significantly. Several studies have reported that the percentage of patients who show clinical improvement after shunting is low (10–61%) and that the complication rate is excessively high (38%) with death or severe residual morbidity in 6–8% of shunted patients.5,14 For all these reasons, predictors are needed identified in order to preoperatively predict the course of the disease. Nevertheless, even today there is no agreement on the best diagnostic criteria for selecting patients with NPH for CSF shunting. In a recent study, standardised questionnaires were sent to 82 neurosurgical centres known to participate in the care of patients with NPH, the result is striking: diagnostic CSF removal was used with varying frequency by all centres except one, but the amount of CSF removed by lumbar puncture differed markedly between them; also wide variability still exists among different neurosurgical centres on the best diagnostic test for selecting patients for shunt surgery.15 Diagnostic tools for patients with suspected NPH are numerous (Table 1).3–8,10,12,14–38 Some of these require specialized equipment and their efficacy and reliability is not always established. The literature is rich in criticism for infusion tests, specific neuropsychological tests, ICP measurements, MRI investigations and for RIC. The saline infusion test with pressure monitoring has been used to reveal decreased CSF resorptive capacity.17 Data from many studies have revealed that the infusion test as well as specific neuropsychological tests are of little or no value in diagnosing NPH. Patients who should be shunted could not be distinguished from those patients who did not need a shunt by any kind of neuropsychological test.35 Meier et al. found that using the intrathecal infusion test, patients with early stage NPH, showed an improvement after a shunt operation in 65% of cases, and those with advanced stage NPH, in only 50% of cases.39 He confirmed later that outflow resistance as measured by the intrathecal infusion test showed only minimal relevance for outcome.40
Table 1 Proposed tests for diagnosing normal pressure hydrocephalus and predicting the outcome of shunt surgery Specific psychometric tests35 Periventricular hyperintensity on MRI, lucency on CT scan10,12,25,27,36–38 Measurement of CSF flow on MRI22,28,32,33 Radioisotope cisternography3,12,16,23,38 CT cisternography12,15,18,34 Single proton emission CT (SPECT)30 Measurement of regional cerebral blood flow18,31 Intrathecal infusion test with ICP measurements3,10,17,26 Ventricular tap test29 Lumbar CSF tap test3,7,8,20,23,25,34 Lumbar external CSF drainage4–6,14,19,21,24,25,38
733
For some, intracranial pressure monitoring is a useful test and B waves occurring during more than 50% of the monitoring period are associated with a likelihood of successful response to shunting.5,41 Several other authors contend that ICP measurements cannot predict outcome in NPH cases,42 that A or B-waves poorly predict which patients will respond to shunt surgery6 and that a statistically significant relationship is not found between percentage of B waves and outcome.10 It has been reported that among patients who underwent ventriculoperitoneal shunting for the treatment of NPH, measurement of CSF flow through the cerebral aqueduct on MRI did not reliably predict which patients would improve after shunting or the magnitude of improvement.5,43 It has also been stated that it is not useful to consider CSF flow void findings on conventional MRI scans in making the decision to offer shunting to patients with idiopathic NPH.28 Nuclear or CT cisternography has been proposed for the evaluation of NPH. If this test shows ventricular reflux with slow cortical uptake, it reveals disordered CSF reabsorption at the level of the arachnoid villi and may have a role in diagnosing communicating hydrocephalus, but it cannot predict shunt response. Its presence does not reliably predict which patients will have favourable outcomes after shunt surgery; similarly, a negative test does not preclude a favourable response to shunt surgery.12,16,23,25 The results of the present series also support this fact: only 66% of patients with positive RIC improved after the shunt surgery. In fact, after 1993, we did not consider a positive RIC sufficient for shunting, and therefore we required an additional positive RTT or LED. It has been published that diagnostic ventricular CSF removal may be a valuable ancillary test in the selection of patients for shunt surgery and that this ‘ventricular tap test’ has shown much greater sensitivity and specificity in selecting which patients will respond to shunting, which is not surprising given that the test comes closest to simulating the actual VP shunt.29 The RTT is an important diagnositc test because of its simplicity and less invasive nature.16,25 It is generally assumed that of the diagnostic studies, the most reliable result is improvement in clinical symptoms following a lumbar puncture in which CSF is withdrawn,20,34 and Milhorat reported that a favourable response to preoperative CSF removal is the most reliable indicator of surgical outcome in patients with NPH.44 Analysis of data from several studies has demonstrated that LED is an accurate test for predicting outcome after ventricular shunting in patients with NPH.6,19,21,24,25 The results of the present study confirm the accuracy of RTT and LED in selecting patients for shunt surgery, with an improvement rate of 88% and 91%, respectively. In the series of Krauss, with the lumbar tap test used as the diagnostic tool provided a similar improvement of 90%.27 In some studies, false negative predictions were high, with the tap test at 58%,3 and the lumbar external drainage
734
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
at 64% false negative predictions.38 It seems that these tests, in their originally described forms, are not accurate enough to reproduce the conditions of a patient operated upon with VPS. Since the 1980s we have used these simple but useful tests with modifications. First, in RTT, the numbers of days where taps are performed was increased to three consecutive days and at each tap a minimum of 30–40 cc CSF was removed. Second, in LED, the volume of CSF drained daily was a minimum of 150–250 cc and this during three consecutive days. Our series has two further differences from many recent reports. First, our patients had at least two symptoms of the clinical triad. We encounter more and more articles where patients with only gait disturbances have been operated upon: this trend of considering NPH primarily as a treatable gait disorder is as dangerous as to consider it primarily as a treatable dementia, much as it was 20 years ago. The high rate of unfavourable outcomes in these series seems to us inevitable: it is well established that post-surgical improvement is greater in patients with the complete clinical triad compared with those who had presented with incomplete forms of the clinical syndrome.10,14,16,35,45,46 The clinical triad is necessary but insufficient, for two reasons: first, it can be encountered in pathologies other than NPH, such as any form of chronic, communicating hydrocephalus, and even a few non-communicating forms such as aqueduct stenosis. All these patients may present with a similar clinical triad, and they may all be treated with a ventriculoperitoneal shunt.12 Cerebral atrophy and deep white matter ischemia are both far more common causes of the clinical triad than NPH.23 Second, the rate of clinical improvement and the positive predictive value of the complete triad are respectively not more than 61% and 65% in the series of NPH patients.16,37 Therefore, the use of a valuable complementary clinical test is mandatory to obtain better results than 65%. Finally, in all operated patients who did not show objective improvement at the follow-up evaluation or worsening of clinical conditions after a temporary improvement, shunt failure was considered and the shunt system was revised in every patient who consented to revision. Ineffective shunt function is a frequent cause of poor outcome after shunt surgery; one must confirm that the shunt is functioning before the diagnosis of NPH is labelled ‘incorrect’. Williams stated in 1998 that ‘poor clinical outcome occurred in two-thirds of all patients after shunt surgery for NPH, but a potentially treatable cause was found in nearly 80% of shunt systems investigated and clinical recovery occurred in 70% of patients who had shunt revision surgery’.47 In the present series, in 49 cases shunt revision was necessary, and after revision, 30 patients (61%) improved with marked benefit. Hebb stated in his review of 44 articles14 an overall improvement of 59% after shunting and only 29% of prolonged improvement. Our results may suggest that with
some modifications in diagnostic tests as mentioned earlier, and with more rigorous patient selection criteria, these results of shunt surgery can be improved in NPH patients. A criticism of this highly selective attitude can be that we excluded some patients who may have benefited from surgery, but the reason for this policy is again the unacceptably high rate of complications, permanent neurological deficit and death in this group of patients and our intention thereby of minimizing unnecessary operations. Otherwise the outcome could have been 6–8% combined rate of permanent neurological deficit and death as in Hebb’s and Vanneste’s studies.5,14 Alzheimer’s and Binswanger’s disease are common sources of comorbidity in older patients with NPH where they contribute to the clinical impairment associated with this disorder. A number of authors have commented on the comorbidities. ‘For patients accurately diagnosed with NPH, concomitant Alzheimer’s disease pathology does not strongly influence the clinical response to shunt surgery.’27,48 ‘Cortical atrophy does not rule out the possibility of improvement after shunt placement, patients should not be denied surgery solely because of dilated sulci or fissures.’10 Our study seems to confirm both these statements: two patients with Parkinson’s disease, four with Alzheimer’s disease and 14 with different forms of sequelae of cerebrovascular accident are of particular interest, as two, two and eight of them respectively, recovered from their NPH symptoms after shunt surgery. 5. Conclusion The results suggest that both the RTT and the LED can accurately predict the outcome of shunt surgery prior to shunt procedures in patients with suspected NPH. The low costs and the simplicity of these tests compared to the invasiveness and the possibility of serious test-related complications of other tests show their usefulness in managing patients with presumed NPH. It has been shown that good results after shunting can be expected in patients showing improvement after a CSF tap test, so surgery may be based on a positive RTT alone. However, if negative, RTT may be complemented by a 3-day LED which is another reliable predictor of favourable outcome after shunting.4,6,14,19,24,25,38 The presence of PVH or subcortical lacunar infarctions in NPH, although of negative prognostic value, does not predict a poor outcome from shunt surgery and should not be used as exclusion criteria for shunting.27,36 From a surgical point of view, the real problem is not to determine the exact diagnostic label, but to identify the appropriate patient to operate on; RTT or LED, when performed in clinically selected patients, seem to provide this information. Our results show that a high percentage of improvement with a low mortality rate and few complications can be achieved in patients with NPH using a rigorous clinical protocol associated with CT/MRI criteria and particularly with positive RTT or LED test results. Our experience
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
and that of others27 indicate that the percentage of improvement after shunting can be as high as 90%, on the condition that this strict diagnostic and treatment protocol is applied. Radioisotope cisternography, lumbar infusion tests and continuous ICP recording can all show that CSF dynamics is ‘deficient or malfunctioning’; they may even be able to establish that the diagnosis is NPH, but none can confirm directly whether the patient will benefit from surgery. In other words, RTT and LED, instead of answering the question, that is of secondary importance to both surgeon and patient, ‘does the patient have NPH or not’, rather responds directly to the question of primary importance: ‘will the patient benefit from this operation, and should they consequently, undergo surgery’. Therefore, the prediction of post-operative-like results, as if surgery has been performed, becomes an invaluable foresight, an opportunity which cannot be encountered in any other situation within the field of neurosurgery. References 1. Adams RD, Fisher CM, Hakim S, et al. Symptomatic occult hydrocephalus with ‘‘Normal’’ cerebrospinal fluid pressure: a treatable syndrome. N Engl J Med 1965;273:117–26. 2. Arriada-Mendicoa N, Herrera-Guerrero MP, Otero-Siliceo E. Chronic hydrocephaly in adults. A diagnostic and therapeutic challenge. Rev Neurol 2002;34:665–72. 3. Kahlon B, Sundba¨rg G, Rehncrona S. Comparison between the lumbar infusion and CSF tap tests to predict outcome after shunt surgery in suspected normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry 2002;73:721–6. 4. Bret P, Guyotat J, Chazal J. Is normal pressure hydrocephalus a valid concept in 2002? A reappraisal in five questions and proposal for a new designation of the syndrome as ‘chronic hydrocephalus’. J Neurol Neurosurg Psychiatry 2002;73:9–12. 5. Vanneste JA. Diagnosis and management of normal-pressure hydrocephalus. J Neurol 2000;247:5–14. 6. Williams MA, Razumovsky AY, Hanley DF. Comparison of PCSF monitoring and controlled CSF drainage diagnose normal pressure hydrocephalus. Acta Neurochir Suppl. 1998;71:328–30. 7. Takeuchi T, Iwasaki M, Shirata K. Evaluation of lumbar tap test combined with mean cerebral blood flow measurement and electroencephalographic topography on idiopathic normal pressure hydrocephalus patients. No Shinkei Geka 2004;32 (3):247–55. 8. Wikkelso¨ C, Andersson H, Blomstrand C, et al. The clinical effect of lumbar puncture in normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry 1982;45:64–9. 9. De Jong DA, Delwel EJ, Avezaat CJ. Hydrostatic and hydrodynamic considerations in shunted normal pressure hydrocephalus. Acta Neurochir (Wien) 2000;142 (3):241–7. 10. Poca MA, Mataro M, Matarin MM, et al. Is the placement of shunts in patients with idiopathic normal-pressure hydrocephalus worth the risk? Results of a study based on continuous monitoring of intracranial pressure. J Neurosurg. 2004;100:855–66. 11. Dunn L. ‘Normal pressure hydrocephalus’: what’s in a name? J Neurol Neurosurg Psychiatry 2002;73:8. 12. Bradley WG. Normal pressure hydrocephalus: new concepts on etiology and diagnosis. Am J Neuroradiol 2000;21:1586–90. 13. Bradley WG. Normal pressure hydrocephalus and deep white matter ischemia: which is the chicken, and which is the egg? Am J Neuroradiol 2001;22:1638–40. 14. Hebb AO, Cusimano MD. Idiopathic normal pressure hydrocephalus: a systematic review of diagnosis and outcome. Neurosurgery 2001;49:1166–84, discussion 1184–6.
735
15. Krauss JK, Halve B. Normal pressure hydrocephalus: survey on contemporary diagnostic algorithms and therapeutic decision-making in clinical practice. Acta Neurochir (Wien) 2004;146:379–88, discussion 388. 16. Black PM. Idiopathic normal-pressure hydrocephalus. Results of shunting in 62 patients. J Neurosurg 1980;52:371–7. 17. Boon AJ, Tans JT, Delwel EJ, et al. Does CSF outflow resistance predict the response to shunting in patients with normal pressure hydrocephalus? Acta Neurochir Suppl (Wien) 1998;71:331–3. 18. Chang CC, Kuwana N, Ito S, et al. Prediction of effectiveness of shunting in patients with normal pressure hydrocephalus by cerebral blood flow measurement and computed tomography cisternography. Neurol Med Chir (Tokyo) 1999;39:841–5, discussion, 845–6. 19. Chen IH, Huang CI, Liu HC, et al. Effectiveness of shunting in patients with normal pressure hydrocephalus predicted by temporary, controlled-resistance, continuous lumbar drainage: a pilot study. J Neurol Neurosurg Psychiatry 1994;57:1430–2. 20. Damasceno BP, Carelli EF, Honorato DC, et al. The predictive value of cerebrospinal fluid tap-test in normal pressure hydrocephalus. Arq Neuropsiquiatr. 1997;55:179–85. 21. Di Lauro L, Viearini M, Bollati A. The predictive value of 5 days CSF diversion for shunting in normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry 1986;49:842–3. 22. Egeler-Peerdeman SM, Barkhof F, Walchenbach R, et al. Cine phasecontrast MR imaging in normal pressure hydrocephalus patients: relation to surgical outcome. Acta Neurochir Suppl (Wien) 1998;71: 340–2. 23. Fishman RA, Dillon WP. Normal pressure hydrocephalus: new findings and old questions. Am J Neuroradiol 2001;22:1640–1. 24. Haan J, Thomeer RTWM. Predictive value of temporary external lumbar drainage in normal pressure hydrocephalus. Neurosurgery 1988;22:388–91. 25. Ishikawa M. Guideline Committee for Idiopathic Normal Pressure Hydrocephalus, Japanese Society of Normal Pressure Hydrocephalus. Clinical guidelines for idiopathic normal pressure hydrocephalus. Neurol Med Chir (Tokyo) 2004;44:222–3. 26. Kosteljanetz M, Nehen AM, Kaalund J. Cerebrospinal fluid outflow resistance measurements in the selection of patients for shunt surgery in the normal pressure hydrocephalus syndrome. A controlled trial. Acta Neurochir 1990;104:48–53. 27. Krauss JK, Droste DW, Vach W, et al. Cerebrospinal fluid shunting in idiopathic normal-pressure hydrocephalus of the elderly: effect of periventricular and deep white matter lesions. Neurosurgery 1996;39: 292–9, discussion 299–300. 28. Krauss JK, Regel JP, Vach W, et al. Flow void of cerebrospinal fluid in idiopathic normal pressure hydrocephalus of the elderly: can it predict outcome after shunting? Neurosurgery 1997;40:67–73, discussion 73–4. 29. Krauss JK, Regel JP. The predictive value of ventricular CSF removal in normal pressure hydrocephalus. Neurol Res 1997;19:357–60. 30. Larsson A, Arlig A, Bergh AC, et al. Quantitative SPECT cisternography in normal pressure hydrocephalus. Acta Neurol Scand 1994;90:190–6. 31. Larsson A, Bergh AC, Bilting M, et al. Regional cerebral blood flow in normal pressure hydrocephalus: diagnostic and prognostic aspects. Eur J Nucl Med 1994;21:118–23. 32. Luertmer PH, Huston J, Friedman JA, et al. Measurement of cerebrospinal fluid flow at the cerebral aqueduct by use of phasecontrast magnetic resonance imaging: technique validation and utility in diagnosing idiopathic normal pressure hydrocephalus. Neurosurgery 2002;50:534–43, discussion 543–4. 33. Mase M, Yamada K, Banno T, et al. Quantitative analysis of CSF flow dynamics using MRI in normal pressure hydrocephalus. Acta Neurochir Suppl (Wien) 1998;71:350–3. 34. Mori K. Management of idiopathic normal-pressure hydrocephalus: a multiinstitutional study conducted in Japan. J Neurosurg 2001;95: 970–3.
736
K. Kilic et al. / Journal of Clinical Neuroscience 14 (2007) 729–736
35. Savolainen S, Hurskainen H, Paljarvi L, et al. Five-year outcome of normal pressure hydrocephalus with or without a shunt: predictive value of the clinical signs, neuropsychological evaluation and infusion test. Acta Neurochir (Wien) 2002;144:515–23, discussion 523. 36. Tullberg M, Jensen C, Ekholm S, et al. Normal pressure hydrocephalus: Vascular white matter changes on MR images must not exclude patients from shunt surgery. Am J Neuroradiol 2001;22:1665–73. 37. Vanneste J, Augustijn P, Tan WF, et al. Shunting normal pressure hydrocephalus: the predictive value of combined clinical and CT data. J Neurol Neurosurg Psychiatry 1993;56:251–6. 38. Walchenbach R, Geiger E, Thomeer RT, et al. The value of temporary external lumbar CSF drainage in predicting the outcome of shunting on normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry 2002;72:503–6. 39. Meier U, Bartels P. The importance of the intrathecal infusion test in the diagnosis of normal pressure hydrocephalus. J Clin Neurosci 2002;9:260–7. 40. Meier U, Konig A, Miethke C. Predictors of outcome in patients with normal-pressure hydrocephalus. Eur Neurol 2004;51:59–67. 41. Reilly P. In normal pressure hydrocephalus, intracranial pressure monitoring is the only useful test. J Clin Neurosci 2001;8:66–7.
42. Sorteberg A, Eide PK, Fremming AD. A prospective study on the clinical effect of surgical treatment of normal pressure hydrocephalus: the value of hydrodynamic evaluation. Br J Neurosurg 2004;18: 149–57. 43. Dixon GR, Friedman JA, Luetmer PH, et al. Use of cerebrospinal fluid flow rates measured by phase-contrast MR to predict outcome of ventriculoperitoneal shunting for idiopathic normal-pressure hydrocephalus. Mayo Clin Proc 2002;77:509–14. 44. Milhorat TH. Comment. Neurosurgery 1997;40:73–4. 45. Janny P, Colnet P, Veyra A, et al. Hydroce´phalie a` pression normale. Etude pre´ et postope´ratoire de 56 cas. Neurochirurgie 1981;27:89–96. 46. Larsson A, Wikkelso¨ C, Bilting M, et al. Clinical parameters in 74 consecutive patients shunt operated for normal pressure hydrocephalus. Acta Neurol Scand 1991;84:475–82. 47. Williams MA, Razumovsky AY, Hanley DF. Evaluation of shunt function in patients who are never better, or better then worse after shunt surgery for NPH. Acta Neurochir Suppl (Wien) 1998;71: 368–70. 48. Golomb J, Wisoff J, Miller DC, et al. Alzheimer’s disease comorbidity in normal pressure hydrocephalus: prevalence and shunt response. J Neurol Neurosurg Psychiatry 2000;68:778–81.