Cerebrospinal fluid pulse pressure method: a possible substitute for the examination of B waves

J Neurosurg 101:944 950, 2004 Cerebrospinal fluid pulse pressure method: a possible substitute for the examination of B waves NIKLAS LENFELDT, M.SC.E.P., B.M., NINA ANDERSSON, M.SC.E.P., AINA ÅGREN-WILSSON, M.D., A. TOMMY BERGENHEIM, M.D., PH.D., LARS-OWE D. KOSKINEN, M.D., PH.D., ANDERS EKLUND, PH.D., AND JAN MALM, M.D., PH.D. Departments of Clinical Neuroscience, and Biomedical Engineering and Informatics; and Centre for Biomedical Engineering and Physics, University of Umeå, Sweden Object. The appearance of numerous B waves during intracranial pressure (ICP) registration in patients with idiopathic adult hydrocephalus syndrome (IAHS) is considered to predict good outcome after shunt surgery. The aim of this study was to describe which physical parameters of the cerebrospinal fluid (CSF) system B-waves reflect and to find a method that could replace long-term B-wave analysis. Methods. Ten patients with IAHS were subjected to long-term registration of ICP and a lumbar constant-pressure infusion test. The B-wave presence, CSF outflow resistance (R out ), and relative pulse pressure coefficient (RPPC) were assessed using computerized analysis. The RPPC was introduced as a parameter reflecting the joint effect of elastance and pulsatory volume changes on ICP and was determined by relating ICP pulse amplitudes to mean ICP. Conclusions. The B-wave presence on ICP registration correlates strongly with RPPC (r = 0.91, p 0.001, 10 patients) but not with CSF R out. This correlation indicates that B waves like RPPC primarily reflect the ability of the CSF system to reallocate and store liquid rather than absorb it. The RPPC-assessing lumbar short-term CSF pulse pressure method could replace the intracranial long-term B-wave analysis. KEY WORDS intracranial pressure normal-pressure hydrocephalus elastance cerebrospinal fluid outflow resistance relative pulse pressure coefficient T HE pathophysiology of IAHS, 31 or idiopathic normalpressure hydrocephalus, 1 is still an enigma. Today two principally different mechanisms are considered: 1) an altered hydrodynamics of the CSF system; and 2) a parenchymal, possibly ischemic, process. The former is demonstrated by modestly raised ICP, increased CSF R out, 31 and the presence of abnormal B-wave patterns; 13 the latter is indicated by cortical and subcortical decreased blood flow 27 and metabolism 2,42 as well as periventricular white matter lesions. 26 In addition, increased blood pressure is a risk factor. 25 The only viable treatment today is shunt surgery; 43 however, only 70% of patients who receive shunts improve clinically, 32 which probably is due to difficulties in selecting appropriate patients. The selection of suitable patients for shunt surgery is based on a variety of examinations 43 and one involves assessing the functionality of the CSF system. Evaluating the functionality of this system involves measurements of specific hydrodynamic parameters such as ICP, R out, and elastance but also targets a less specific characteristic, namely the presence of B-waves. Abbreviations used in this paper: B% = B-wave percentage; CSF = cerebrospinal fluid; IAHS = idiopathic adult hydrocephalus syndrome; ICP = intracranial pressure; PC = personal computer; PVI = pressure volume index; R out = outflow resistance; RPPC = relative pulse-pressure coefficient. 944 Several CSF hydrodynamic tests are used to assess R out. The basic concept is to study the pressure-infusion curve through either constant flow infusion 22 or constant pressure infusion. 18 Elastance is determined by using bolus injections to measure the PVI of the CSF pressure volume curve, 33 where PVI is the volume needed to increase the pressure 10-fold. Other methods measure how ICP is influenced by the combined effect of elastance and volume change. Examples include the CSF pulse pressure method 3 and the ICP waveform analysis, 16 which both make use of the pulsatory changes in the intracranial blood volume and their transformation into pressure waves in the CSF system. The B-waves are slow and rhythmic oscillations in ICP. 29 They probably originate from oscillations in cerebral blood volume 36 and may be present in healthy individuals as well. 35 Examination of B waves is invasive, involving surgical insertion of a probe followed by pressure registration lasting for several hours. In general, the results of the registration are evaluated visually by clinicians, but recently we have described a computerized method of calculating the B-wave presence of an ICP recording. 17 The various methods used to measure the hydrodynamic properties of the CSF system have not been properly compared. Consequently, the internal relations between the aforementioned parameters are unknown. If the presence of B-waves could be explained by other parameters, it would

Cerebrospinal fluid pulse pressure and B waves Fig. 1. Upper: Schematic of current circuit model, once described by Marmarou, illustrating the hydrodynamics of the CSF system. Currents in the model symbolize flows in the CSF system. When ICP exceeds dural venous pressure (P D ), the switch closes and a CSF absorption flow (I A ) occurs through a non pressure-dependent resistor representing R out. The CSF storage flow (I S ) that is, the CSF volume rate of change is associated with the elastance of the CSF system, and the storage capability is represented by a pressure-dependent capacitor. The production of CSF generates a one-way production flow (I P ). Similarly, the ability to withdraw or add CSF/mock CSF produces a two-way infusion flow (I I ). To consider volume oscillations of the intracranial blood compartment, the model is extended by an alternating current (I B ) that represents the blood volume change rate. These oscillations, including pulse waves and B waves, are transferred to the CSF system and affect the distribution of CSF. Lower: Graph illustrating that the relative distribution between storage flow (solid line) and absorption flow (dashed line) in the CSF system depends on the wavelength (T) of the imposed volume oscillation. Shorter waves, such as pulse waves (T 1 second log T = 0) and B-waves (T range of 30 120 seconds log T range of 1.5 2.1), lie in the storage zone; that is, almost 100% of these volume oscillations are dealt with by reallocation and storage of CSF, whereas longer waves of 30 minutes (T = 1800 seconds log T = 3.3) are needed to create an evenly distributed flow. Infinite wavelength, which is the approximate situation during a constant infusion test, yields a 100% absorption flow, and the test can therefore be used to determine R out. The curves for the flows are calculated using the circuit analysis of the model in Fig. 1 upper. See the Appendix for details. be possible to replace overnight ICP recording and B-wave analysis with simpler methods to select patients likely to benefit from shunt placement. In this study, we compared three methods based on an extended model of the CSF system (Fig. 1 upper). Our aim was to ascertain how the B-wave pattern in patients with IAHS is related to R out, which was measured using the constant pressure infusion method, and to the newly defined parameter of RPPC, which was derived from the CSF pulse pressure method. Patient Population Clinical Material and Methods The population in the current prospective study was part of a larger series of patients with IAHS. This larger series included two main CSF hydrodynamic studies: an overnight ICP registration and a constant pressure infusion test. Microdialysis and monitoring of brain tissue PO 2 were also performed. The local ethics committee approved all aspects of both the larger series and the smaller study and informed 945

N. Lenfeldt, et al. FIG. 2. Case 4. Graph demonstrating ICP registration from the constant pressure infusion in a patient. Note the relaxation phase used in the CSF pulse pressure method to determine the RPPC. consent was obtained from each patient prior to the hydrodynamic examinations. This study was based on 10 patients eight men and two women with typical radiological signs of IAHS, that is, communicating hydrocephalus with no signs of major ischemic lesions, severe cortical atrophy, or widespread white matter lesions. Clinically, all patients showed the characteristic gait disturbance (a compulsory sign that was recorded by video) often accompanied with mental decline and/or incontinence. The mean patient age was 72.4 years (range 55 78 years) and the mean duration of gait disturbance was 28 months (range 12 60 months). After finishing all examinations, all patients were offered a shunt based on the comprehensive picture of their clinical, radiological, and hydrodynamic data. All patients agreed to undergo shunt placement and 1 to 2 months after the examinations each of them received a CSF shunt device. At the 3- to 6-month follow up after shunt surgery, nine of 10 patients (Cases 2 10) experienced gait improvement. Assessment of B Waves The steps involved in overnight ICP registration and B-wave analysis have previously been described. 17 In brief, an intraparenchymal ICP transducer (Codman microsensor; Johnson & Johnson Professional, Inc., Raynham, MA) was inserted in the morning. The ICP registration started at approximately noon on the same day and was finished at 8 a.m. the following morning (range 17.5 20.8 hours). The patient was lying in bed, awake or asleep, during the entire period of data collection. The sampling rate was 100 Hz and data were recorded on a PC (Power PC 7600; Apple Computer, Inc., Cupertino, CA) by using a multimodal recording system including a data acquisition card (MIO16X50; National Instruments, Inc., Austin, TX) and commercially available software (LabVIEW; National Instruments). The 100-Hz data were averaged over 1 second, giving an actual sampling rate of 1 Hz, and were analyzed using a program developed in LabVIEW and solely designed for the purpose of B-wave analysis. The program filtered the ICP data through a digital bandpass filter of the Butterworth type (5th order), which cut off frequencies outside the B-wave interval of 0.5 to 2 minutes. The B waves were revealed using a computer algorithm for individual wave analysis. A wave was classified as a FIG. 3. Graph demonstrating magnification and simplification of the relaxation phase featured in Fig. 2. The CSF pulse pressure method uses this phase to detect the impact of the pulse waves on ICP. This information is then used to determine RPPC. As shown, an increased ICP is followed by an increased ICP amplitude. The RPPC determines the magnitude of this increase and is derived in the following manner for several ICPs: the minimal (P min ) and maximal (P max ) pressures of the ICP pulse wave are registered, and the ICP amplitude is calculated as P = P max P min. The P is related to a reference pressure, that is, mean ICP, defined as P mean = P min P/3. 3 The values of P and P mean are plotted against each other and a linear regression analysis is performed. The RPPC is determined as the slope (k) of the linear regression equation: P = k P mean m (see the example in Fig. 4). B wave if the absolute value of its peak or trough equaled a minimal threshold value of 1 mm Hg. A wave was defined as a new wave when the data passed the zero level. The B% was calculated as the accumulated time with intervals classified as B waves divided by the total time. Determination of CSF R out Determination of R out was made using the previously described method of constant pressure infusion. 30 In short, at 8 a.m. on the second day, directly after the B-wave examination, two needles were inserted into the lumbar subarachnoid space while the patient was sitting upright. Free passage was assessed and the patient was then placed supine. Drainage and infusion of real or artificial CSF were performed using a peristaltic pump (MS-1 REGLO=160; Ismatec SA, Zurich, Switzerland) through one of the lumbar needles. Using the other lumbar needle, data acquisition and pressure regulation were performed with the aid of an electronic control unit developed at Umeå University and LabVIEW software run on a PC with a data acquisition board (MacADIOS; GW Instruments, Inc., Somerville, MA). Pressure measurements were conducted using an ICP monitoring kit (Becton, Dickinson, and Company, Franklin Lakes, NJ) calibrated at 0 and 22.5 mm Hg. The zero pressure level was calibrated in the cranial sagittal center, which was defined as the midpoint between the highest and lowest point of the body placed supine. Outflow resistance was determined by applying three pressure levels (two in patients in Cases 1 3) above resting pressure to the CSF space (Fig. 2), while recording the required rate of artificial CSF inflow. The first applied pressure was at least 7.5 mm Hg above resting pressure, where- 946

Cerebrospinal fluid pulse pressure and B waves TABLE 1 Results of B-wave presence, R out, and RPPC analyses in 10 patients* Case No. B% R out (mm Hg min/ml) RPPC (dimensionless) FIG. 4. Case 4. Plot of P compared with P mean in a patient. Performing a linear regression analysis yields RPPC = 0.75 (r = 0.96, p 0.0001, 300 samples). as the last and highest level was always 41.5 to 45 mm Hg. When an intermediate pressure level was used, it was applied halfway between the first and last (highest) level. The lower levels were shorter in time (10 15 minutes), whereas the top level lasted for 20 to 30 minutes. The last level was terminated by halting the pump, and ICP returned to its resting pressure. Finally, the CSF system was drained before the test was finished. Corresponding flow/pressure values were plotted, and the slope, which was equal to the inverse of R out, was determined using linear regression. Assessing RPPC by the CSF Pulse Pressure Method The basic principle of the CSF pulse pressure method is to measure the pulsatory ICP amplitudes at different ICPs. 3 The application of the method and its necessary definitions are explained in Fig. 3. The derivation of the RPPC is also described. To detect ICP amplitudes during constant pressure infusion, pressure data from the ICP monitoring kit were continuously sampled from the lumbar subarachnoid space at a frequency of 100 Hz. Data were conveyed to a PC by an experimental setup that was identical to the one used in the B-wave examination. The ICP amplitude can be adequately determined during the relaxation phase of the constant pressure infusion (Fig. 2) because the pump is at rest and cannot interfere with the natural shape of the amplitude. In addition, ICP passes through a vast pressure interval while relaxing toward its resting pressure. The relaxation phase started when the pump was turned off at the end of the last pressure level and lasted for a period of 10 minutes. The period was divided into 2-second intervals, which ensured that one heartbeat was captured within each interval. Pressure data from each interval were then analyzed using a computer algorithm developed in LabVIEW. The computer algorithm was designed to single out the maximal and minimal pressures in each interval, yielding a total of 300 pairs of maximal/minimal values. Pressure trends were ignored. We defined the RPPC to describe quantitatively the change in ICP amplitude during the relaxation phase. The RPPC was determined for each patient according to the method featured in Fig. 3 and an example is shown in Fig. 4. 1 82.3 12.05 0.97 2 67.45 11.40 0.85 3 61.83 16.49 0.75 4 47.68 8.47 0.69 5 41.1 15.67 0.58 6 50.24 16.71 0.57 7 75.68 33.88 1.03 8 68.83 27.25 0.69 9 45.07 15.10 0.64 10 88.55 36.87 1.07 mean 62.87 19.39 0.78 SD 16.46 9.79 0.19 * SD = standard deviation. The coefficient of correlation of each RPPC value was below 0.9 only in the patients in Cases 4 and 5 (r = 0.84), whereas the probability value was less than 0.0001 in all 10 patients. Statistical Analysis The B-wave percentage, R out, and RPPC were compared using linear regression: B% = c 1 RPPC m 1, (1a) B% = c 2 R out m 2, and (1b) R out = c 3 RPPC m 3, (1c) where c i is the regression coefficient and m 1 is the interception of the regression line with the dependent variable axis. To further investigate the relative importance of RPPC and R out to describe B%, two additional tests were conducted. First, the residuals (Res) of Equation 1a were plotted against R out and a linear regression was performed: Res Eq. 1 = c 4 R out m 4. (2) Second, R out was added to Equation 1a as an explanatory variable and a forward multiple regression was performed: B% = c 5 RPPC c 6 R out m 5. (3) The coefficient of correlation and probability value were determined for the regression coefficients and the level of significance was set at 0.05. Results The results from the B-wave analysis (B%), constant pressure infusion (R out ), and RPPC are presented in Table 1. The relationship between B% and RPPC shows a significant correlation (r = 0.91, p 0.001, 10 patients), whereas no significant correlation was found between either B% and R out (r = 0.59, p = 0.07, 10 patients) or R out and RPPC (r = 0.55, p = 0.10, 10 patients). The residual analysis did not show a correlation (r = 0.22, p = 0.54, 10 patients); hence, resistance cannot account for what is left to explain after using the linear B% RPPC model. A forward multiple regression demonstrated that adding R out as an explanatory variable does not yield a significantly better correlation compared with the linear B% RPPC regression. Discussion Despite the B waves unknown relation to other param- 947

N. Lenfeldt, et al. eters of the CSF system, their presence in ICP recordings has been used to select patients with IAHS for shunt surgery 9,13,39 ever since Lundberg defined them. 29 Our data demonstrate a strong correlation between the presence of B-waves and our defined RPPC, indicating the possibility of replacing the former with the latter to select shunt-responsive patients with IAHS. The RPPC, Elastance, and B Waves The linear regression between P and P mean (Fig. 4), defining the RPPC, showed a high correlation in each individual case (Table 1). Similar analyses performed in animals 4,6 and humans 3,15,16,40 demonstrate high correlations as well, despite differences in methodology. This result indicates a robust method. The RPPC and elastance are not the same, which would explain why previous attempts to correlate B-waves with elastance have failed. 10,38,44 Elastance determines the pressure volume curve, 33 whereas RPPC is determined by the combined effect of elastance and volume change, the latter of which results from intracranial blood pulsations. 3 Thus, RPPC can be interpreted as an individually scaled parameter that reflects the total strain put on the CSF system by pulsating blood flow. Previous work has revealed weak correlations between elastance and RPPC, 3,5,6,37 which probably is due to the variability in intracranial pulsation volume among individuals. Cases of stronger correlations have been reported, 24 but any comparison would be tenuous because the RPPC was defined differently. The RPPC, R out, and B Waves Our results indicate a correlation between RPPC and R out but not a significant one; a similar relationship was found between B waves and R out. In contrast, the correlation between B% and RPPC was highly significant. This piece of evidence along with the results from the residual analysis and the multiple regression thoroughly supports the assertion that RPPC alone is the parameter that most accurately indicates the B-wave presence. In the literature, both weaker 10,11,38 and stronger 9,14,19 correlations between B waves and R out have been reported. In addition, B waves have been found to correlate exponentially with the sum of R out and the inverse of elastance. 10,38 Any differences may depend on methodological issues, 7,12,20,23,30,33 such as variability in methods of measuring R out and subjectivity in the definition 11,36,44 and visual interpretation of B waves. Nonhomogeneous groups of patients, 19 arising from erroneous or indefinite IAHS diagnosis may contribute to the inconsistencies as well. Results of our residual analysis and multiple regression demonstrate that the correlation between B waves and R out could be secondary to the correlation between R out and RPPC. The B Waves and RPPC: an Explanatory Model The close link between B waves and RPPC can be explained by taking into consideration an extended model of the CSF system (Fig. 1 upper). Pulse waves and most likely B waves originate from intracranial blood volume oscillations (I B in Fig. 1 upper). These oscillations are then transferred to the CSF system, in agreement with the Monroe Kellie doctrine. The CSF system can deal with imposed oscillations in two separate ways: either reallocate them within the system, thereby imitating CSF storage (storage flow, I S in Fig. 1 upper), or divert them out of the system by absorbing CSF (absorption flow, I A in Fig. 1 upper) into the sinuses. The distribution between the two that is, how much is stored and how much is absorbed depends on the wavelength of the blood oscillation, ICP, PVI (that is, elastance), and R out (Fig. 1 lower). By using common numbers in the context of IAHS, it can be shown (Fig. 1 lower) that the CSF system primarily uses storage, not absorption, to deal with both pulse waves and B waves. Thus, despite a much longer wavelength, B waves reflect the same functional portion of the CSF system as the RPPC, namely its ability to temporarily reallocate and store supplementary intracranial volumes. Idiopathic Adult Hydrocephalus Syndrome, B Waves, and Shunt Surgery Researchers at centers with a long practice in IAHS maintain that approximately 70% of patients improve after shunt surgery. 8,28,32 From this perspective, the 90% improvement rate in our study reflects a correct initial diagnosis in our patients. In the hands of some, 9,13,21,39 the usefulness of B-wave analysis to predict response from shunt surgery is substantiated. Others 17 assert that more solid scientific evidence is needed before we can speak knowingly of the precision of B waves in corroborating IAHS and predicting shunt responsiveness. Today, there is no conclusive proof that either B-wave examination or other methods, such as R out measurements, magnetic resonance imaging, or tap tests, can accurately forecast the benefit of a shunt in a general case of IAHS. The methods are likely to be complementary. Hence, there is a great need to develop a more reliable test to select patients with IAHS for shunt surgery, especially in cases in which the diagnosis is indefinite. In agreement with the described pattern of inconclusiveness, this study revealed no distinct deviation in B-wave presence or R out in a comparison between the nine patients whose condition improved and the one patient whose condition did not. Regarding accessibility to the method, note that B-wave analysis is limited to centers where ICP monitoring can be performed, 43 thus demanding facilities that offer both neurosurgical procedures and pressure registration. Measuring R out or using the CSF pulse pressure method is less complex, requiring lumbar pressure regulation only, and magnetic resonance imaging and spinal tap tests are widely available methods. Conclusions Our data demonstrate a strong correlation between RPPC and the presence of B waves. This result indicates that B waves like RPPC primarily reflect the functionality of reallocation and storage rather than absorption in the CSF system. The CSF pulse pressure method has obvious advantages. By adopting the same lumbar infusion technique that determines R out, one can assess the same hydrodynamics in 10 minutes as those revealed in a B-wave examination lasting several hours. Furthermore, the method is not limited to use 948

Cerebrospinal fluid pulse pressure and B waves in patients with IAHS and may be useful in other cases of intracranial monitoring as well. Using an objective B-wave analysis and a homogeneous patient population enhanced the reliability of our results, and we conclusively believe that the RPPC measured with the CSF pulse pressure method may replace the B-wave analysis as a means of selecting patients with IAHS who will benefit from shunt placement. Appendix By applying circuit analysis to the model in Fig. 1 upper, the ratio between CSF storage and absorption flow is calculated as follows: I S /I A = (2 f C)/(1/R out ), (4a) where f is the wave frequency and C is the compliance, that is, the inverse of elastance. If C = 1/(k P), k = 1/(0.4343 PVI), 34 P = ICP, and T = 1/f then I S /I A = (R out PVI 2 0.4343/ICP)/T, (4b) where T is the wavelength. More specifically, if values of R out and pressure from earlier IAHS data are inserted (R out = 15.1 mm Hg min/ml, ICP = 16.6 mm Hg, 30 ) and the PVI of hydrocephalus material = 17.5 ml, 41 then I S /I A = 2600/T, (4c) where T is measured in seconds. To determine what proportion of blood volume oscillation (I B ) is distributed as storage flow (I S ) and absorption flow (I A ), the two latter are divided by the former, yielding I S /I B = I S /(I S I A ) = 2600/(2600 T) (5a) for the storage flow and I A /I B = I A /(I S I A ) = T/(2600 T) (5b) for the absorptive flow. Plotting Equation 5a and b against the logarithm of T generates the graph featured in Fig. 1 lower. 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 273:117 126, 1965 2. Agren-Wilsson A, Roslin M, Eklund A, et al: Intracerebral microdialysis and CSF hydrodynamics in idiopathic adult hydrocephalus syndrome. J Neurol Neurosurg Psychiatry 74: 217 221, 2003 3. Avezaat CJ, van Eijndhoven JH: Clinical observations on the relationship between cerebrospinal fluid pulse pressure and intracranial pressure. Acta Neurochir 79:13 29, 1986 4. Avezaat CJ, van Eijndhoven JH: The role of the pulsatile pressure variations in intracranial pressure monitoring. Neurosurg Rev 9: 113 120, 1986 5. Avezaat CJ, van Eijndhoven JH, de Jong DA, et al: A new method of monitoring intracranial volume-pressure relationships, in Beks JWF, Bosch DA, Brock M (eds): Intracranial Presure III. Berlin: Springer-Verlag, 1976, pp 308 313 6. Avezaat CJ, van Eijndhoven JH, Wyper DJ: Cerebrospinal fluid pulse pressure and intracranial volume-pressure relationships. J Neurol Neurosurg Psychiatry 42:687 700, 1979 7. Boon AJ, Tans JT, Delwel EJ, et al: Dutch normal-pressure hydrocephalus study: prediction of outcome after shunting by resistance to outflow of cerebrospinal fluid. J Neurosurg 87:687 693, 1997 8. Borgesen SE: Conductance to outflow of CSF in normal pressure hydrocephalus. Acta Neurochir 71:1 45, 1984 9. Borgesen SE, Gjerris F: The predictive value of conductance to outflow of CSF in normal pressure hydrocephalus. Brain 105: 65 86, 1982 10. Borgesen SE, Gjerris F, Sorensen SC: Cerebrospinal fluid conductance and compliance of the craniospinal space in normalpressure hydrocephalus. A comparison between two methods for measuring conductance to outflow. J Neurosurg 51:521 525, 1979 11. Borgesen SE, Gjerris F, Sorensen SC: Intracranial pressure and conductance to outflow of cerebrospinal fluid in normal-pressure hydrocephalus. J Neurosurg 50:489 493, 1979 12. Borgesen SE, Gjerris F, Srensen SC: The resistance to cerebrospinal fluid absorption in humans. A method of evaluation by lumbo-ventricular perfusion, with particular reference to normal pressure hydrocephalus. Acta Neurol Scand 57:88 96, 1978 13. Crockard HA, Hanlon K, Duda EE, et al: Hydrocephalus as a cause of dementia: evaluation by computerised tomography and intracranial pressure monitoring. J Neurol Neurosurg Psychiatry 40:736 740, 1977 14. Czepko R: Znaczenie Kliniczne fal B w zapisie cisnienia plynumozgowurdzeniowego podczas doledzwiowego testu infuzysnego u chorych z wodoglowiem. Przeglad Lekarski 56: 568 575, 1999 15. Czosnyka M, Guazzo E, Whitehouse M, et al: Significance of intracranial pressure waveform analysis after head injury. Acta Neurochir 138:531 542, 1996 16. Czosnyka M, Wollk-Laniewski P, Batorski L, et al: Analysis of intracranial pressure waveform during infusion test. Acta Neurochir 93:140 145, 1988 17. Eklund A, Agren-Wilsson A, Andersson N, et al: Two computerized methods used to analyze intracranial pressure B waves: comparison with traditional visual interpretation. J Neurosurg 94: 392 396, 2001 18. Ekstedt J: CSF hydrodynamic studies in man. 1. Method of constant pressure CSF infusion. J Neurol Neurosurg Psychiatry 40: 105 119, 1977 19. Gjerris F, Borgesen SE: Current concepts of measurement of cerebrospinal fluid absorption and biomechanics of hydrocephalus. Adv Techn Stand Neurosurg 19:145 177, 1992 20. Gjerris F, Borgesen SE, Hoppe E, et al: The conductance to outflow of CSF in adults with high-pressure hydrocephalus. Acta Neurochir 64:59 67, 1982 21. Graff-Radford NR, Godersky JC, Jones MP: Variables predicting surgical outcome in symptomatic hydrocephalus in the elderly. Neurology 39:1601 1604, 1989 22. Katzman R, Hussey F: A simple constant-infusion manometric test for measurement of CSF absorption. I. Rationale and method. Neurology 20:534 544, 1970 23. Kosteljanetz M: CSF dynamics in patients with subarachnoid and/ or intraventricular hemorrhage. J Neurosurg 60:940 946, 1984 24. Kosteljanetz M: Pressure-volume conditions in patients with subarachnoid and/or intraventricular hemorrhage. J Neurosurg 63: 398 403, 1985 25. Krauss JK, Regel JP, Vach W, et al: Vascular risk factors and arteriosclerotic disease in idiopathic normal-pressure hydrocephalus of the elderly. Stroke 27:24 29, 1996 26. Krauss JK, Regel JP, Vach W, et al: White matter lesions in patients with idiopathic normal pressure hydrocephalus and in an age-matched control group: a comparative study. Neurosurgery 40:491 496, 1997 27. Kristensen B, Malm J, Fagerland M, et al: Regional cerebral blood flow, white matter abnormalities, and cerebrospinal fluid hydrodynamics in patients with idiopathic adult hydrocephalus syndrome. J Neurol Neurosurg Psychiatry 60:282 288, 1996 28. Larsson A, Wikkelso C, Bilting M, et al: Clinical parameters in 74 consecutive patients shunt operated for normal pressure hydrocephalus. Acta Neurol Scand 84:475 482, 1991 29. Lundberg N: Continuous recording and control of ventricular fluid pressure in neurosurgical practice. Acta Psychiatr Scand 36 (Suppl 149):1 193, 1960 30. Lundkvist B, Eklund A, Kristensen B, et al: Cerebrospinal fluid hydrodynamics after placement of a shunt with an antisiphon device: a long-term study. J Neurosurg 94:750 756, 2001 949

N. Lenfeldt, et al. 31. Malm J, Kristensen B, Karlsson T, et al: The predictive value of cerebrospinal fluid dynamic tests in patients with idiopathic adult hydrocephalus syndrome. Arch Neurol 52:783 789, 1995 32. Malm J, Kristensen B, Stegmayr B, et al: Three-year survival and functional outcome of patients with idiopathic adult hydrocephalus syndrome. Neurology 55:576 578, 2000 33. Marmarou A, Shulman K, LaMorgese J: Compartmental analysis of compliance and outflow resistance of the cerebrospinal fluid system. J Neurosurg 43:523 534, 1975 34. Marmarou A, Shulman K, Rosende RM: A nonlinear analysis of the cerebrospinal fluid system and intracranial pressure dynamics. J Neurosurg 48:332 344, 1978 35. Mautner-Huppert D, Haberl RL, Dirnagl U, et al: B-waves in healthy persons. Neurol Res 11:194 196, 1989 36. Newell DW, Aaslid R, Stooss R, et al: The relationship of blood flow velocity fluctuations to intracranial pressure B waves. J Neurosurg 76:415 421, 1992 37. Rieger A, Ebel H, Laun A: Korrelation zwischen der Steigungskonstanten des theoretischen Druckvolumendiagramms (PVD) und des Druckvolumenindex (PVT) in relation zur Hirnpulsamplituden-Mitte/druckfunktion. Zentralbl Neurochir 56: 43 48, 1995 38. Sahuquillo J, Rubio E, Codina A, et al: Reappraisal of the intracranial pressure and cerebrospinal fluid dynamics in patients with the so-called normal pressure hydrocephalus syndrome. Acta Neurochir 112:50 61, 1991 39. Symon L, Dorsch NW: Use of long-term intracranial pressure measurement to assess hydrocephalic patients prior to shunt surgery. J Neurosurg 42:258 273, 1975 40. Szewczykowski J, Sliwka S, Kunicki A, et al: A fast method of estimating the elastance of the intracranial system. J Neurosurg 47: 19 26, 1977 41. Tans JT, Poortvliet DC: CSF outflow resistance and pressure-volume index determined by steady-state and bolus infusions. Clin Neurol Neurosurg 87:159 165, 1985 42. Tedeschi E, Hasselbalch SG, Waldemar G, et al: Heterogeneous cerebral glucose metabolism in normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry 59:608 615, 1995 43. Vanneste JA: Diagnosis and management of normal-pressure hydrocephalus. J Neurol 247:5 14, 2000 44. Wilkinson HA, Schuman N, Ruggiero J: Nonvolumetric methods of detecting impaired intracranial compliance or reactivity: pulse width and wave form analysis. J Neurosurg 50:758 767, 1979 Manuscript received March 12, 2004. Accepted in final form July 30, 2004. The study was supported by Karl-Oskar Hansson s Foundation; King Gustav V s and Queen Victoria s Foundation; The Swedish Society of Neurologically Disabled (NHR); Alzheimer Foundation, Sweden; Demensförbundet (The Dementia Association, Sweden); The Foundation for Clinical Neuroscience at Umeå University Hospital; and Umeå University. Address reprint requests to: Niklas Lenfeldt, M.Sc., B.M., Department of Clinical Neuroscience, Umeå University Hospital, S-901 85 Umeå, Sweden. email: niklas.lenfeldt@neuro.umu.se. 950