Original Research Paper
Cerebrovascular time constant in patients suffering from hydrocephalus Cyrille Capel1,2, Magdalena Kasprowicz3, Marek Czosnyka1, Olivier Baledent2, Piotr Smielewski1, John D. Pickard1, Zofia Czosnyka1 1
Academic Neurosurgical Unit, University of Cambridge, Addenbrooke’s Hospital, UK, 2BioFlow Image Department, Hospital University Center, Amiens, France, 3Institute of Biomedical Engineering and Instrumentation, Wroclaw University of Technology, Poland
Objectives: We studied possible link between cerebrospinal fluid (CSF) compensation and indices describing pulsatile inflow of cerebral arterial blood. Methods: A total of 50 infusion tests performed in patients with symptoms of normal pressure hydrocephalus (NPH) were examined retrospectively. Waveforms of CSF pressure, noninvasive arterial blood pressure (ABP), and transcranial Doppler (TCD) cerebral blood flow velocity (CBFV) were used to estimate relative changes in cerebral arterial compliance (Ca) and cerebrovascular resistance (CVR). Product of Ca and CVR, called cerebral arterial time constant (t, unit: seconds), was calculated at the baseline and plateau phase of the test and compared with CSF compensatory parameters such as resistance to CSF outflow, elasticity, slope of amplitude–pressure line, and pulse amplitude of CSF pressure. Results: Neither of CSF compensatory parameters correlated with hemodynamic indices. However, the change in cerebral perfusion pressure (CPP) provoked change in t (R 5 0.33; P 5 0.017) secondary to a change in CVR (R 5 0.81; P , 0.0001). Changes in CVR and Ca had a reciprocal character (R 5 20.64; P , 0.0001) with magnitude of variation in CVR (68%) prevailing over magnitude of changes in Ca (49%). Discussion: Hemodynamics of pulsatile inflow of cerebral arterial blood assessed by cerebral arterial time constant is not directly linked to dynamics of CSF circulation and pressure–volume compensation but is sensitive to changes in CPP during infusion test. Keywords: Hydrocephalus, Cerebrospinal fluid, Cerebral blood flow velocity, Infusion test
Introduction Although adult hydrocephalus is a well-recognized syndrome, not all of its underlying physiopathological phenomena are understood in depth. Many hypotheses were proposed historically, but majority of neuroscientists agree that hydrocephalus is initiated by an inadequate relationship between secretion, circulation, and reabsorption of cerebrospinal fluid (CSF).1–3 Cerebrospinal fluid infusion study to measure the resistance of CSF outflow is a routine procedure to diagnose patients with ventricular dilation and symptoms such as gait disturbance, cognitive trouble, and urinary incontinence.1 Although increased resistance to CSF outflow is correlated with better chance for improvement after shunting,2,3 some recent studies challenge this thesis.4,28 Moreover, overall improvement depends on crosscombination of multiple factors: pre-shunting CSF Correspondence to: Marek Czosnyka, Neurosurgery Unit, Box 167, Addenbrooke’s Hospital, Cambridge CB2 0QQ, UK. Email: [email protected]. ac.uk
ß W. S. Maney & Son Ltd 2014 DOI 10.1179/1743132813Y.0000000282
circulation profile, shunt function after implantation, and the coexistence of other diseases including cerebrovascular pathologies.3,5 Using a multimodality recording during the infusion study (intracranial pressure (ICP), arterial blood pressure (ABP), and transcranial Doppler (TCD) cerebral blood flow velocity (CBFV)), we previously have demonstrated that the resistance to CSF outflow correlates with a state of autoregulation of cerebral blood flow (CBF).6 Disturbed autoregulation allows detection of underlying cerebrovascular disease and predicts difficulty in the process of improving after shunt surgery. Recently, a method to monitor changes in compartmental brain compliances was developed.7,8 The recent study in animal model9 analyzed a scenario of hemodynamic changes during the infusion study. Controlled elevation of ICP produced a decrease in CPP associated with a significant increase in the arterial compliance of the large cerebral arteries (Ca) and a fall in cerebrovascular resistance (CVR) of the
Neurological Research
2014
VOL .
36
NO .
3
255
Capel et al.
Cerebrovascular time constant in hydrocephalus
small regulatory vessels.9 Therefore, we can anticipate that if hydrocephalus patients have a preserved autoregulation, during the infusion study we could see a vasodilatation of distal territory vessels and an increase in the value of compliance (Ca). Cerebrovascular resistance changes in the same direction as CPP and Ca in the opposite direction. Measurement of absolute values of Ca and CVR requires a knowledge of constant cross-sectional area of examined (insonated) vessel. This prerequisite is difficult with TCD techniques. A recently introduced cerebral arterial time constant10 can characterize cerebral hemodynamics during the infusion study. It investigates the relationship between Ca and CVR. This ultrasound-based index is independent of the vessel’s cross-sectional area and has physically interpreted units (seconds). Theoretically, t describes how fast the arterial bed distal to the point of insonation (place of CBFV measurement) is filled with blood,9–11 following the heart constriction. In this study, we investigated changes in t during the infusion test in patients diagnosed for hydrocephalus.
Material and Methods Patients and monitoring A total of 50 patients (median age: 56.5 years, range: 22–78) were retrospectively selected from a larger database of patients diagnosed in the direction of normal pressure hydrocephalus (NPH) who underwent a constant-rate infusion study in Addenbrooke’s Hospital from 1994 to 2006. The criterion for selection was a clear, without artifacts, simultaneous recording of ABP, ICP, and TCD CBFV performed during the infusion study. All patients had clinical symptoms including gait disturbance, cognitive disorders, and urinary incontinence. All of them had a degree of ventricular dilation seen on CT or MRI scan. They were examined as a part of routine preshunting investigation in Hydrocephalus Clinic of Addenbrooke’s Hospital (JDP), and results are analyzed and presented here as a part of routine clinical audit in anonymous ways. We used a CSF computerized infusion test, allowing a computer-aided analysis of the traditional constant rate infusion into CSF space.12 This methodology has been published previously.13–15 The test was performed using an infusion into subcutaneous reservoir connected to an intraventricular catheter. Two lines were used with two needles (gage 25). First line allowed the pressure measurement by connection to a pressure transducer via a stiff saline-filled tube. The second line was connected to an infusion pump and allowed the control of the infusion with a constant rate of 1.5 or 1 ml/minute (when the baseline pressure was above 15 mmHg). A
256
Neurological Research
2014
VOL .
36
NO .
3
purpose-built trolley contains the infusion pump, an amplifier (Simonsen & Will Sidcup, UK), and an IBM compatible laptop running ICMz (Neurosurgery Unit, University of Cambridge, http:// www.neurosurg.cam.ac.uk/icmplus).16 Recording of ABP using Finapres plethysmograph (Finapres Medical Systems B.V., Amsterdam, The Netherlands) and CBFV in MCA using TCD (Neuroguard, Medasonics, Cremona, CA, USA) were performed along the infusion study. Cerebral blood flow velocity was recorded bilaterally in 43 cases. After 10 minutes of baseline measurement, the infusion of Hartman’s solution at a constant rate was started. The infusion was continued till the ICP plateau was achieved. If ICP persistently increased above 40 mmHg, infusion was stopped.
Data analysis From ICP recorded during the infusion study, CSF compensatory parameters were assessed: baseline ICP, baseline amplitude of pulse wave, resistance to CSF outflow, and elasticity. Resistance to CSF outflow describes circulation of CSF,12 whereas elasticity, pressure/volume compensation.15 These are standard parameters derived from infusion studies. Changes in cerebral blood volume during a cardiac cycle (t0 was start of cardiac cycle) could be calculated as an integral of the difference between the pulsatile arterial inflow and the venous outflow:7 ðt DCBV~ (CBFa (t){CBFv (t))dt (1) t0
where CBFa is the cerebral arterial blood inflow, CBFv the cerebral venous blood outflow, and DCBV the changes in cerebral blood volume during one cardiac cycle. Venous outflow has a lower pulsatility than arterial inflow; therefore, it can be approximated by the averaged arterial inflow: DCa BV(n)~S a
n X
ðCBFVa (i){meanCBFVa ÞDt
(2)
i~1
where Sa is the cross-sectional area of insonated vessel, n the number of samples from the start of a cardiac cycle, Dt the interval between two samples, CBFVa the changes in cerebral blood velocity, and mean CBFVa the average of CBFVa. In equation (2), CBFa was substituted by CBFVa multiplied by cross-sectional area of insonated artery (Sa). Sa is unknown and we assumed that it remains constant during one cardiac cycle. We applied Fourier transformation to calculate the pulse change amplitude of fundamental harmonics of both ABP and DCaBV. We estimated Ca [cm3/ mmHg] as the quotient of CaBV amplitude by ABP amplitude.
Capel et al.
Ca ~AMPCaBV =AMPABP
(3)
CVR [mmHg/(cm3/s)] was the mean CPP divided by CBFV. CVR~
CPP CBFV:Sa
(4)
Ca is a function of Sa (see equations (2) and (3)) and CVR is a function of inversed Sa (equation (4)). Consequently, the product of Ca and CVR cancels the unknown parameter Sa from calculations. The product of Ca and CVR is called the cerebral arterial time constant (t) and is expressed in seconds. t~C a :CVR
(5)
Cerebral arterial time constant can be interpreted as the time of filling the distal arterial bed to the point of insonation of the vessel.9–11
Statistical analysis Paired sign rank test was used to compare the differences in median values of analyzed parameters (ABP, ICP, CBFV, t, Ca, CVR) between the baseline and plateau phase. Because CVR and Ca are functions of unknown cross-sectional area of insonated vessel (Sa) and cannot be calibrated in [cm3/ mmHg] and [mmHg *s/cm3], respectively, their percentage changes during infusion test were compared between patients. Association between selected variables was investigated using Spearman correlation. The level of significance was set at , 0.05. Data were reported as median (lower and upper quartile).
Results Cerebrospinal fluid compensatory parameters are listed in Table 1. Median baseline ICP was 8.7 mmHg. A total of 34% of patients had resistance to CSF outflow (Rcsf) elevated above 13 mmHg/[ml/minute]. Resistance to CSF outflow correlated positively with baseline ICP (R 5 0.35; P , 0.015) and baseline amplitude of CSF pressure pulse waveform (R 5 0.42; P , 0.003). When compared patients with normal and elevated Rcsf (. 13 mmHg/[ml/minute] – possible
Cerebrovascular time constant in hydrocephalus
responders to shunt implantation), no differences have been found between these two groups either in percentage changes in CVR or Ca or in absolute changes in t. There were also no correlations between Rcsf and above-listed parameters. Typical infusion study with rising ICP, decreasing CPP, slightly decreasing CBFV, decreasing CVR, very slightly increasing Ca (only in initial period in this case), and slightly decreasing t is presented in Fig. 1. Values of pressures, CBFV, Ca, CVR, and t before and during plateau phase of the study are presented in Table 2. There was no correlation between any of CSF compensatory parameters and t, Ca, or CVR, neither at baseline nor at plateau phase of infusion. In most of tests, an increase in ICP during the study was associated with a decrease in CPP (R5 20.5; P 5 0.0002). On average, ABP increased slightly during the study (P , 0.0005). In 10 tests, a rise in arterial pressure was greater than an increase in ICP; therefore, CPP increased during these studies. When compared those patients who had a drop in CPP and those who had an increase in CPP during the infusion test, the significant difference between percentage changes in CVR (8.84 (215.34; 21.25)% vs 10.28 (9.50; 20.56)%, P , 0.00001) and Ca (21.99 (216.03; 8.28)% vs 15.29 (6.39; 29.66)%, P 5 0.001) has been found but not in absolute changes in t [s]. Ca and CVR changed during infusion studies and these changes had a mutually reciprocal character – see Fig. 2 (R 5 20.64; P , 0.0001), with magnitude of variation in CVR (68%) prevailing over magnitude of changes in Ca (49%). Variations in t were correlated with variations in CPP (but not with variations in ICP) during the infusion study – see Fig. 3C. Cerebrovascular resistance decreased strongly with the decrease in CPP (R 5 0.81; P , 0.0001, Fig. 3B). Ca increased moderately with decreasing CPP (R520.37; P 5 0.008, Fig. 3A). As a result, t followed changes in CVR, correlating weakly with decreasing CPP (R 5 0.33; P 5 0.017). Cerebrovascular resistance was inversely related to changes in ICP (R 5 20.29; P 5 0.04), whereas Ca was independent of changes in ICP. Changes in ABP correlated strongly with CVR (R 5
Table 1 Cerebrospinal fluid (CSF) compensatory parameters. Results are presented as median, lower quartile (Q1), and upper quartile (Q2) Parameter ICP baseline [mmHg] ICP pulse amplitude at baseline [mmHg] Resistance to CSF outflow [mmHg/[ml/minute] Elasticity [1/ml] Sagittal sinus pressure [mmHg] CSF formation rate [ml/minute] Slope of amplitude/pressure line
Median
Q1; Q3
8.70 1.66 9.95 0.19 3.22 0.31 0.22
2.64; 11.2 0.86; 2.28 7.32; 14.07 0.1; 0.38 20.9; 8.21 0.11; 0.71 0.14; 0.31
ICP: intracranial pressure.
Neurological Research
2014
VOL .
36
NO .
3
257
Capel et al.
Cerebrovascular time constant in hydrocephalus
Figure 1 Typical time trends of 10-second averaged variables recorded and calculated using ICMz during the infusion study. ICP: mean intracranial pressure (CSF pressure in ventricles, recorded from Ommaya reservoir); ABP: arterial blood pressure; CPP: cerebral perfusion pressure (ABP-ICP); CBFV: MCA blood flow velocity; Ca: compliance of cerebral arterial bed; CVR: cerebrovascular resistance (both CVR and Ca values are presented per unit of cross-sectional area of MCA [cm2]); t: cerebral arterial time constant; and x-axis: time in hours and minutes. Infusion with a rate of 1.5 ml/minute started at the first vertical marker and finished at the second marker.
0.64; P , 0.000001) and weakly with Ca (R 5 20.3; P 5 0.03).
Discussion Cerebrospinal fluid dynamics compared to arterial hemodynamic indices Cerebral arterial time constant is a new parameter describing cerebral arterial bed. It theoretically describes time needed for arterial blood after each heart stroke to arrive at hypothetical arteriole– capillary border in brain. It may be expressed in seconds or in a fraction of heart cycle (average in volunteers is 20–30%10). Previously we demonstrated that time constant shortens in hypercapnia in volunteers10 and prolongs (experimentally) when CPP decreases.9 In disease, it has been reported that time constant becomes shorter in carotid artery
stenotic disease17 and in cerebral vasospasm after subarachnoid haemorrhage.11 In the present study, we wanted to investigate potential links between CSF compensatory parameters and cerebral hemodynamic indices associated with t. We also wanted to investigate changes in cerebral hemodynamics during the infusion study. Previously, we described a relationship between cerebral autoregulation and resistance to CSF outflow.6 Patients with normal CSF circulation tend to have more frequently disturbed autoregulation, which probably reflects an existing cerebrovascular disease. However, time constant describes different properties of brain’s vasculature – associated with fast blood transport (fraction of seconds) rather than slow (seconds or even tens of seconds) regulation of vascular tone. Neither of indices (t, CVR, Ca)
Table 2 Comparison between baseline and plateau phase of infusion test (50 patients). Results are presented as median, lower quartile (Q1), and upper quartile (Q2) Median (Q1; Q3) baseline ICP [mmHg] ABP [mmHg] CPP [mmHg] CBFV [cm/s] CVR [mmHg/(cm/s)] Ca [cm/mmHg] t [s]
8.7 (2.6; 11.2) 89.9 (77.8; 103.3) 81.1 (72.6; 94.5) 48.3 (41.6; 62.4) 1.81 (1.28; 2.03) 0.11 (0.08; 0.16) 0.20 (0.14; 0.24)
Median (Q1; Q3) plateau 20.3 94.0 76.0 45.6 1.65 0.11 0.18
(16.6; 27.2) (84.3; 119.1) (63.0; 97.8) (37.25; 60.0) (1.34; 2.16) (0.82; 0.15) (0.14; 0.23)
Paired sign rank test 5.32610210 0.000003 0.0002 4.6361027 0.045 0.76 0.014
ICP: intracranial pressure; ABP: arterial blood pressure; CPP: cerebral perfusion pressure; CBFV: cerebral blood flow velocity; CVR: cerebrovascular resistance; Ca: cerebral arterial compliance (both CVR and Ca values are presented per unit of cross-sectional area of MCA [cm2]); t: cerebrovascular time constant.
258
Neurological Research
2014
VOL .
36
NO .
3
Capel et al.
Cerebrovascular time constant in hydrocephalus
Figure 2 Reciprocal character of percentage changes in cerebral arterial compliance (Ca) and cerebrovascular resistance (CVR) during infusion test. Variation in CVR (68%) prevails over changes in Ca (49%).
correlated with any of CSF compensatory parameters. Cerebral hemodynamic indices describe conditions to arterial blood inflow, CSF compensatory parameter conditions for CSF flow – which has a rate almost 2000 times lower than CBF. Our previous study18 indicated that Ca does not change during infusion test in contrast to decreasing cerebrospinal compartmental compliance. But cerebrovascular time constant has been never considered in hydrocephalus scenario. Infusion test modifies t, which gets significantly shorter. This shortening is a result of a significant decrease in CVR and non-significant increase in Ca. The decrease in t and increase in Ca are dependent on changes in CPP (Fig. 3A and C) but are independent of changes in ICP during the infusion test. The increase in ICP during infusion could not have been strong enough to activate changes in Ca, or/and the reactivity of cerebral arterial compliance to ICP changes might have been weakened by ongoing pathological process. In result, t is not correlated with changes in ICP within the investigated range of pressure variations. Ca and CVR usually change in opposite directions8 (see Fig. 2). During the infusion study, a decrease in CPP causes gradual vasodilatation (decrease in CVR) and an increase in cerebral arterial compliance (due to relaxation of arterial-arterioral smooth muscle). In experimental conditions,9 we found that an increase in Ca is deeper than a decrease in CVR; therefore t becomes longer. In hydrocephalus, a decrease in CVR prevails over an increase in Ca and t becomes shorter (swing of CVR is 68%, whereas Ca is only 49%). In experimental rabbits, the magnitude of a decrease in CPP during the infusion test was bigger than in patients with NPH. In fact, in 10 patients CPP increased during the infusion study, whereas such a situation did not have place in rabbits. Furthermore, rabbits were anesthetized, whereas
Figure 3 Relationships between changes induced by infusion in cerebral perfusion pressure (CPP [mmHg]) and percentage change in (A) cerebral arterial compliance (Ca [%]), (B) cerebrovascular resistance (CVR [%]), and (C) absolute changes in cerebral arterial time constant (t [s]). Decrease in CPP during infusion causes a decrease in CVR due to a vasodilatation of distal vessels and an increase in Ca. Changes in CVR prevail over changes in Ca; therefore, t varies in the same direction as CVR.
patients were conscious. Anesthetics may have influenced vessels’ tone and their reactivity. Finally, the reactivity of compliance may have been affected (more probably – at larger diameter vessels) by the process of hydrocephalus. It remains to be investigated whether inversed balance between changes in Ca and CVR in normal rabbits and hydrocephalic patients has anything in common with patophysiology of hydrocephalus.
Neurological Research
2014
VOL .
36
NO .
3
259
Capel et al.
Cerebrovascular time constant in hydrocephalus
Cerebral perfusion pressure and cerebral arterial time constant: the influence of vascular reactivity During the infusion test, CPP on average decreased even if 10 patients presented a slight increase in this parameter, due to a rise in ABP previously described as the early Cushing response.19 As a result of CPP decreasing, we observed a decreasing CVR. Conversely, with increasing CPP, CVR increased (see Fig. 3B). Decreasing of CVR indicates a vasodilatation of distal vessels allowing to maintain a constant CBF. This may indicate that vascular reactivity is in general preserved in NPH patients.
Arterial remodeling effect In a recent study,20 the authors found a relation between arterial stiffness and CVR, which confirms that increasing Ca is correlated with decreasing CVR. In another study,21 the authors showed that atherosclerosis and CVR could be affected by different etiologies, as activity of antioxidant enzymes. It could explain the inverse correlation between Ca and CVR. Arterial remodeling, arterial wall calcifications, and atherosclerosis can affect Ca and CVR, independently of CPP. Another mechanism can be explored after our preliminary work. Idiopathic orthostatic intolerance patients present a decreasing MCA flow velocity response to head-up tilt. This effect may be explained as impairment of cerebrovascular regulation.22 Cerebral arterial time constant depends physiologically on CPP and PaCO2 and in pathology – on many factors responsible or associated with developing disease. Cerebral arterial time constant is not dependant on cross-sectional area of insonated vessel and also independent of angle of insonation of the vessel. t derivates from the time constant of electric resistor–capacitor circuit.9–11 It is the interval required for a circuit or physiological system to switch from one state to another.
Limitations Calculations on cerebral arterial compliance (Ca) assumed that the venous outflow is constant during a cardiac cycle. Phase-contrast MRI study23 showed pulsatility of venous outflow in superior sagittal sinus (SSS) and jugular vein. Pulsatility increased from SSS to jugular vein. This pulsatility is lower than arterial inflow. Cross-sectional area of the insonated large vessel is not constant. It consistently changed Ca and CVR, but t is independent of this area. Finally, ABP was recorded from a peripheral artery with the pulsatile blood pressure (fundamental harmonic). It is not an ideal approximation of cerebral ABP.24 Our thesis that cerebral arterial time constant changes during the infusion study because of predominantly intact cerebrovascular reactivity, i.e.
260
Neurological Research
2014
VOL .
36
NO .
3
positive association between CVR and CPP, does not exhaust all questions about possible links between the nature of the disease and regulation of blood flow. We previously demonstrated possible disturbance of autoregulation in white matter.25 There is documented association between autoregulation and hormonal activity26 and autonomic drive.27 It is possible that the results of the study are interfered by inaccuracy of method of the cerebral arterial time constant estimation, and a link between compensatory parameters and cerebral hemodynamic indices might exist. Currently, TCD ultrasonography is the only method that allows for continuous measurement of changes in cerebral hemodynamics during the infusion study. Imaging techniques, although more accurate, can provide only a snapshot of the state of cerebral hemodynamics and their use during the infusion study is difficult due to technical limitations.
Conclusion The cerebrovascular time constant changes during the infusion study mainly due to autoregulatory CVR response to changing CPP. t is not dependent on CSF dynamics in NPH.
Conflict of Interest ICMz is a clinical neuroscience software supporting brain monitoring. It is licensed by University of Cambridge, Cambridge Enterprise Ltd (www.neurosurg.cam.ac.uk/icmplus). PS and MC have a financial interest in a fraction of licensing fee.
Acknowledgements Many thanks to Drs Christina Haubrich and Ruwan Weerakkody for assisting in collection of clinical data and all colleagues working for Hydrocephalus Clinic for their help, friendly advice, and clinical data collection. MK was supported by the Polish Ministry of Science and Higher Education.
References 1 Hakim S, Adams RD. The special clinical problem of symptomatic hydrocephalus with normal cerebrospinal fluid pressure: observations on cerebrospinal fluid hydrodynamics. J Neurol Sci. 1965;2:307–27. 2 Borgesen SE, Gjerris F. The predictive value of conductance to outflow of CSF in normal pressure hydrocephalus. Brain. 1982;105:65–86. 3 Boon AJ, Tans JT, Delwel EJ, Egeler-Peerdeman SM, Hanlo PW, Wurzer HA, et al. Dutch normal-pressure hydrocephalus study: the role of cerebrovascular disease. J Neurosurg. 1999;90:221–6. 4 Eide PK, Brean A. Cerebrospinal fluid pulse pressure amplitude during lumbar infusion in idiopathic normal pressure hydrocephalus can predict response to shunting. Cerebrospinal Fluid Res. 2010;7:5. 5 Kristensen B, Malm J, Fagerlund M, Hietala SO, Johanson B, Ekstedt J, et al. Regional cerebral blood flow, white matter abnormalities, and cerebrospinal fluid hydrodynamics in patients with idiopathic adult hydrocephalus syndrome. J Neurol Neurosurg Psychiatry. 1996;60:282–8.
Capel et al.
6 Czosnyka Z, Czosnyka M, Whitfield P, Donovan T, Pickard J. Cerebral autoregulation among patients with symptoms of hydrocephalus. Neurosurgery. 2002;50:526–33. 7 Czosnyka M, Richards HK, Reinhard M, Steiner LA, Budohoski K, Smielewski P, et al. Cerebrovascular time constant: dependence on cerebral perfusion pressure and endtidal carbon dioxide concentration. Neurol Res. 2012;34(1):17– 24. 8 Kim DJ, Kasprowicz M, Carrera E, Castellani G, Zweifel C, Lavinio A, et al. The monitoring of relative changes in compartmental compliances of brain. Physiol. Meas. 2009;30:647–59. 9 Carrera E, Kim DJ, Castellani G, Zweifel C, Smielewski P, Pickard JD, et al. Effect of hyper- and hypocapnia on cerebral arterial compliance in normal subjects. J Neuroimaging. 2011;21:121–5. 10 Kasprowicz M, Diedler J, Reinhard M, Carrera E, Steiner LA, Smielewski P, et al. Time constant of the cerebral arterial bed in normal subjects. Ultrasound Med Biol. 2012;38(7):1129–37. 11 Kasprowicz M, Czosnyka M, Soehle M, Smielewski P, Kirkpatrick PJ, Pickard JD, et al. Vasospasm shortens cerebral arterial time constant. Neurocrit Care. 2012 Apr;16(2):213–8. 12 Børgesen SE, Albeck MJ, Gjerris F, Czosnyka M, Laniewski P. Computerized infusion test compared to steady pressure constant infusion test in measurement of resistance to CSF outflow. Acta Neurochir (Wien). 1992;119:12–6. 13 Czosnyka M, Pickard JD. Monitoring and interpretation of intracranial pressure. J Neurol Neurosurg Psychiatry. 2004;75:813–21. 14 Czosnyka M, Wollk-Laniewski P, Batorski L, Zaworski W. Analysis of intracranial pressure waveform during infusion test. Acta Neurochir (Wien). 1988;93:140–5. 15 Weerakkody RA, Czosnyka M, Schuhmann MU, Schmidt E, Keong N, Santarius T, et al. Clinical assessment of cerebrospinal fluid dynamics in hydrocephalus. Guide to interpretation based on observational study. Acta Neurol Scand. 2011;124(2):85–98. 16 Smielewski P, Lavinio A, Timofeev I, Radolovich D, Perkes I, Pickard JD, et al. ICMz, a flexible platform for investigations of cerebrospinal dynamics in clinical practice. Acta Neurochir Suppl. 2008;102:145–51. 17 Kasprowicz M, Diedler J, Reinhard M, Carrera E, Smielewski P, Budohoski KP, et al. Time constant of the cerebral arterial bed. Acta Neurochir Suppl. 2012;114:17–21.
Cerebrovascular time constant in hydrocephalus
18 Kim DJ, Carrera E, Czosnyka M, Keong N, Smielewski P, Bale´dent O, et al. Cerebrospinal compensation of pulsating cerebral blood volume in hydrocephalus. Neurol Res. 2010;32(6):587–92. 19 Schmidt EA, Czosnyka Z, Momjian S, Czosnyka M, Bech RA, Pickard JD. Intracranial baroreflex yielding an early Cushing response in human. Acta Neurochir Suppl. 2005;95:253–6. 20 Robertson AD, Tessmer CF, Hughson RL. Association between arterial stiffness and cerebrovascular resistance in the elderly. J Hum Hypertens. 2010;24:190–6. 21 D’Armiento FP, Bianchi A, de Nigris F, Capuzzi DM, D’Armiento MR, Crimi G, et al. Age-related effects on atherogenesis and Scavenger enzymes of intracranial and extracranial arteries in men without classic risk factors for atherosclerosis. Stroke. 2001;32:2472–80. 22 Jacob G, Atkinson D, Jordan J, Shannon JR, Furlan R, Black BK, et al. Effects of standing on cerebrovascular resistance in patient with idiopathic orthostatic intolerance. Am J Med. 1999;106:59–64. 23 Stoquart-ElSankari S, Lehmann P, Villette A, Czosnyka M, Meyer M-E, Deramond H, et al. A phase-contrast MRI study of physiologic cerebral venous flow. J Cereb Blood Flow Metab. 2009;29:1208–15. 24 O’Rourke MF, Adji A. Noninvasive studies of central aortic pressure. Curr Hypertens Rep. 2012;14(1):8–20. 25 Momjian S, Owler BK, Czosnyka Z, Czosnyka M, Pena A, Pickard JD. Pattern of white matter regional cerebral blood flow and autoregulation in normal pressure hydrocephalus. Brain. 2004;127(5):965–72. 26 Hong KW, Shin HK, Kim CD, Lee WS, Rhim BY. Restoration of vasodilation and CBF autoregulation by genistein in rat pial artery after brain injury. Am J Physiol Heart Circ Physiol. 2001;281(1):308–15. 27 Lavinio A, Ene-Iordache B, Nodari I, Girardini A, Cagnazzi E, Rasulo F, et al. Cerebrovascular reactivity and autonomic drive following traumatic brain injury. Acta Neurochir Suppl. 2008;102:3–7. 28 Wikkelsø C, Hellstro¨m P, Klinge PM, Tans JT; European iNPH Multicentre Study Group. The European iNPH Multicentre Study on the predictive values of resistance to CSF outflow and the CSF Tap Test in patients with idiopathic normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry. 2013 May;84(5):562–8.
Neurological Research
2014
VOL .
36
NO .
3
261
Cerebrovascular time constant in patients suffering from hydrocephalus Cyrille Capel1,2, Magdalena Kasprowicz3, Marek Czosnyka1, Olivier Baledent2, Piotr Smielewski1, John D. Pickard1, Zofia Czosnyka1 1
Academic Neurosurgical Unit, University of Cambridge, Addenbrooke’s Hospital, UK, 2BioFlow Image Department, Hospital University Center, Amiens, France, 3Institute of Biomedical Engineering and Instrumentation, Wroclaw University of Technology, Poland
Objectives: We studied possible link between cerebrospinal fluid (CSF) compensation and indices describing pulsatile inflow of cerebral arterial blood. Methods: A total of 50 infusion tests performed in patients with symptoms of normal pressure hydrocephalus (NPH) were examined retrospectively. Waveforms of CSF pressure, noninvasive arterial blood pressure (ABP), and transcranial Doppler (TCD) cerebral blood flow velocity (CBFV) were used to estimate relative changes in cerebral arterial compliance (Ca) and cerebrovascular resistance (CVR). Product of Ca and CVR, called cerebral arterial time constant (t, unit: seconds), was calculated at the baseline and plateau phase of the test and compared with CSF compensatory parameters such as resistance to CSF outflow, elasticity, slope of amplitude–pressure line, and pulse amplitude of CSF pressure. Results: Neither of CSF compensatory parameters correlated with hemodynamic indices. However, the change in cerebral perfusion pressure (CPP) provoked change in t (R 5 0.33; P 5 0.017) secondary to a change in CVR (R 5 0.81; P , 0.0001). Changes in CVR and Ca had a reciprocal character (R 5 20.64; P , 0.0001) with magnitude of variation in CVR (68%) prevailing over magnitude of changes in Ca (49%). Discussion: Hemodynamics of pulsatile inflow of cerebral arterial blood assessed by cerebral arterial time constant is not directly linked to dynamics of CSF circulation and pressure–volume compensation but is sensitive to changes in CPP during infusion test. Keywords: Hydrocephalus, Cerebrospinal fluid, Cerebral blood flow velocity, Infusion test
Introduction Although adult hydrocephalus is a well-recognized syndrome, not all of its underlying physiopathological phenomena are understood in depth. Many hypotheses were proposed historically, but majority of neuroscientists agree that hydrocephalus is initiated by an inadequate relationship between secretion, circulation, and reabsorption of cerebrospinal fluid (CSF).1–3 Cerebrospinal fluid infusion study to measure the resistance of CSF outflow is a routine procedure to diagnose patients with ventricular dilation and symptoms such as gait disturbance, cognitive trouble, and urinary incontinence.1 Although increased resistance to CSF outflow is correlated with better chance for improvement after shunting,2,3 some recent studies challenge this thesis.4,28 Moreover, overall improvement depends on crosscombination of multiple factors: pre-shunting CSF Correspondence to: Marek Czosnyka, Neurosurgery Unit, Box 167, Addenbrooke’s Hospital, Cambridge CB2 0QQ, UK. Email: [email protected]. ac.uk
ß W. S. Maney & Son Ltd 2014 DOI 10.1179/1743132813Y.0000000282
circulation profile, shunt function after implantation, and the coexistence of other diseases including cerebrovascular pathologies.3,5 Using a multimodality recording during the infusion study (intracranial pressure (ICP), arterial blood pressure (ABP), and transcranial Doppler (TCD) cerebral blood flow velocity (CBFV)), we previously have demonstrated that the resistance to CSF outflow correlates with a state of autoregulation of cerebral blood flow (CBF).6 Disturbed autoregulation allows detection of underlying cerebrovascular disease and predicts difficulty in the process of improving after shunt surgery. Recently, a method to monitor changes in compartmental brain compliances was developed.7,8 The recent study in animal model9 analyzed a scenario of hemodynamic changes during the infusion study. Controlled elevation of ICP produced a decrease in CPP associated with a significant increase in the arterial compliance of the large cerebral arteries (Ca) and a fall in cerebrovascular resistance (CVR) of the
Neurological Research
2014
VOL .
36
NO .
3
255
Capel et al.
Cerebrovascular time constant in hydrocephalus
small regulatory vessels.9 Therefore, we can anticipate that if hydrocephalus patients have a preserved autoregulation, during the infusion study we could see a vasodilatation of distal territory vessels and an increase in the value of compliance (Ca). Cerebrovascular resistance changes in the same direction as CPP and Ca in the opposite direction. Measurement of absolute values of Ca and CVR requires a knowledge of constant cross-sectional area of examined (insonated) vessel. This prerequisite is difficult with TCD techniques. A recently introduced cerebral arterial time constant10 can characterize cerebral hemodynamics during the infusion study. It investigates the relationship between Ca and CVR. This ultrasound-based index is independent of the vessel’s cross-sectional area and has physically interpreted units (seconds). Theoretically, t describes how fast the arterial bed distal to the point of insonation (place of CBFV measurement) is filled with blood,9–11 following the heart constriction. In this study, we investigated changes in t during the infusion test in patients diagnosed for hydrocephalus.
Material and Methods Patients and monitoring A total of 50 patients (median age: 56.5 years, range: 22–78) were retrospectively selected from a larger database of patients diagnosed in the direction of normal pressure hydrocephalus (NPH) who underwent a constant-rate infusion study in Addenbrooke’s Hospital from 1994 to 2006. The criterion for selection was a clear, without artifacts, simultaneous recording of ABP, ICP, and TCD CBFV performed during the infusion study. All patients had clinical symptoms including gait disturbance, cognitive disorders, and urinary incontinence. All of them had a degree of ventricular dilation seen on CT or MRI scan. They were examined as a part of routine preshunting investigation in Hydrocephalus Clinic of Addenbrooke’s Hospital (JDP), and results are analyzed and presented here as a part of routine clinical audit in anonymous ways. We used a CSF computerized infusion test, allowing a computer-aided analysis of the traditional constant rate infusion into CSF space.12 This methodology has been published previously.13–15 The test was performed using an infusion into subcutaneous reservoir connected to an intraventricular catheter. Two lines were used with two needles (gage 25). First line allowed the pressure measurement by connection to a pressure transducer via a stiff saline-filled tube. The second line was connected to an infusion pump and allowed the control of the infusion with a constant rate of 1.5 or 1 ml/minute (when the baseline pressure was above 15 mmHg). A
256
Neurological Research
2014
VOL .
36
NO .
3
purpose-built trolley contains the infusion pump, an amplifier (Simonsen & Will Sidcup, UK), and an IBM compatible laptop running ICMz (Neurosurgery Unit, University of Cambridge, http:// www.neurosurg.cam.ac.uk/icmplus).16 Recording of ABP using Finapres plethysmograph (Finapres Medical Systems B.V., Amsterdam, The Netherlands) and CBFV in MCA using TCD (Neuroguard, Medasonics, Cremona, CA, USA) were performed along the infusion study. Cerebral blood flow velocity was recorded bilaterally in 43 cases. After 10 minutes of baseline measurement, the infusion of Hartman’s solution at a constant rate was started. The infusion was continued till the ICP plateau was achieved. If ICP persistently increased above 40 mmHg, infusion was stopped.
Data analysis From ICP recorded during the infusion study, CSF compensatory parameters were assessed: baseline ICP, baseline amplitude of pulse wave, resistance to CSF outflow, and elasticity. Resistance to CSF outflow describes circulation of CSF,12 whereas elasticity, pressure/volume compensation.15 These are standard parameters derived from infusion studies. Changes in cerebral blood volume during a cardiac cycle (t0 was start of cardiac cycle) could be calculated as an integral of the difference between the pulsatile arterial inflow and the venous outflow:7 ðt DCBV~ (CBFa (t){CBFv (t))dt (1) t0
where CBFa is the cerebral arterial blood inflow, CBFv the cerebral venous blood outflow, and DCBV the changes in cerebral blood volume during one cardiac cycle. Venous outflow has a lower pulsatility than arterial inflow; therefore, it can be approximated by the averaged arterial inflow: DCa BV(n)~S a
n X
ðCBFVa (i){meanCBFVa ÞDt
(2)
i~1
where Sa is the cross-sectional area of insonated vessel, n the number of samples from the start of a cardiac cycle, Dt the interval between two samples, CBFVa the changes in cerebral blood velocity, and mean CBFVa the average of CBFVa. In equation (2), CBFa was substituted by CBFVa multiplied by cross-sectional area of insonated artery (Sa). Sa is unknown and we assumed that it remains constant during one cardiac cycle. We applied Fourier transformation to calculate the pulse change amplitude of fundamental harmonics of both ABP and DCaBV. We estimated Ca [cm3/ mmHg] as the quotient of CaBV amplitude by ABP amplitude.
Capel et al.
Ca ~AMPCaBV =AMPABP
(3)
CVR [mmHg/(cm3/s)] was the mean CPP divided by CBFV. CVR~
CPP CBFV:Sa
(4)
Ca is a function of Sa (see equations (2) and (3)) and CVR is a function of inversed Sa (equation (4)). Consequently, the product of Ca and CVR cancels the unknown parameter Sa from calculations. The product of Ca and CVR is called the cerebral arterial time constant (t) and is expressed in seconds. t~C a :CVR
(5)
Cerebral arterial time constant can be interpreted as the time of filling the distal arterial bed to the point of insonation of the vessel.9–11
Statistical analysis Paired sign rank test was used to compare the differences in median values of analyzed parameters (ABP, ICP, CBFV, t, Ca, CVR) between the baseline and plateau phase. Because CVR and Ca are functions of unknown cross-sectional area of insonated vessel (Sa) and cannot be calibrated in [cm3/ mmHg] and [mmHg *s/cm3], respectively, their percentage changes during infusion test were compared between patients. Association between selected variables was investigated using Spearman correlation. The level of significance was set at , 0.05. Data were reported as median (lower and upper quartile).
Results Cerebrospinal fluid compensatory parameters are listed in Table 1. Median baseline ICP was 8.7 mmHg. A total of 34% of patients had resistance to CSF outflow (Rcsf) elevated above 13 mmHg/[ml/minute]. Resistance to CSF outflow correlated positively with baseline ICP (R 5 0.35; P , 0.015) and baseline amplitude of CSF pressure pulse waveform (R 5 0.42; P , 0.003). When compared patients with normal and elevated Rcsf (. 13 mmHg/[ml/minute] – possible
Cerebrovascular time constant in hydrocephalus
responders to shunt implantation), no differences have been found between these two groups either in percentage changes in CVR or Ca or in absolute changes in t. There were also no correlations between Rcsf and above-listed parameters. Typical infusion study with rising ICP, decreasing CPP, slightly decreasing CBFV, decreasing CVR, very slightly increasing Ca (only in initial period in this case), and slightly decreasing t is presented in Fig. 1. Values of pressures, CBFV, Ca, CVR, and t before and during plateau phase of the study are presented in Table 2. There was no correlation between any of CSF compensatory parameters and t, Ca, or CVR, neither at baseline nor at plateau phase of infusion. In most of tests, an increase in ICP during the study was associated with a decrease in CPP (R5 20.5; P 5 0.0002). On average, ABP increased slightly during the study (P , 0.0005). In 10 tests, a rise in arterial pressure was greater than an increase in ICP; therefore, CPP increased during these studies. When compared those patients who had a drop in CPP and those who had an increase in CPP during the infusion test, the significant difference between percentage changes in CVR (8.84 (215.34; 21.25)% vs 10.28 (9.50; 20.56)%, P , 0.00001) and Ca (21.99 (216.03; 8.28)% vs 15.29 (6.39; 29.66)%, P 5 0.001) has been found but not in absolute changes in t [s]. Ca and CVR changed during infusion studies and these changes had a mutually reciprocal character – see Fig. 2 (R 5 20.64; P , 0.0001), with magnitude of variation in CVR (68%) prevailing over magnitude of changes in Ca (49%). Variations in t were correlated with variations in CPP (but not with variations in ICP) during the infusion study – see Fig. 3C. Cerebrovascular resistance decreased strongly with the decrease in CPP (R 5 0.81; P , 0.0001, Fig. 3B). Ca increased moderately with decreasing CPP (R520.37; P 5 0.008, Fig. 3A). As a result, t followed changes in CVR, correlating weakly with decreasing CPP (R 5 0.33; P 5 0.017). Cerebrovascular resistance was inversely related to changes in ICP (R 5 20.29; P 5 0.04), whereas Ca was independent of changes in ICP. Changes in ABP correlated strongly with CVR (R 5
Table 1 Cerebrospinal fluid (CSF) compensatory parameters. Results are presented as median, lower quartile (Q1), and upper quartile (Q2) Parameter ICP baseline [mmHg] ICP pulse amplitude at baseline [mmHg] Resistance to CSF outflow [mmHg/[ml/minute] Elasticity [1/ml] Sagittal sinus pressure [mmHg] CSF formation rate [ml/minute] Slope of amplitude/pressure line
Median
Q1; Q3
8.70 1.66 9.95 0.19 3.22 0.31 0.22
2.64; 11.2 0.86; 2.28 7.32; 14.07 0.1; 0.38 20.9; 8.21 0.11; 0.71 0.14; 0.31
ICP: intracranial pressure.
Neurological Research
2014
VOL .
36
NO .
3
257
Capel et al.
Cerebrovascular time constant in hydrocephalus
Figure 1 Typical time trends of 10-second averaged variables recorded and calculated using ICMz during the infusion study. ICP: mean intracranial pressure (CSF pressure in ventricles, recorded from Ommaya reservoir); ABP: arterial blood pressure; CPP: cerebral perfusion pressure (ABP-ICP); CBFV: MCA blood flow velocity; Ca: compliance of cerebral arterial bed; CVR: cerebrovascular resistance (both CVR and Ca values are presented per unit of cross-sectional area of MCA [cm2]); t: cerebral arterial time constant; and x-axis: time in hours and minutes. Infusion with a rate of 1.5 ml/minute started at the first vertical marker and finished at the second marker.
0.64; P , 0.000001) and weakly with Ca (R 5 20.3; P 5 0.03).
Discussion Cerebrospinal fluid dynamics compared to arterial hemodynamic indices Cerebral arterial time constant is a new parameter describing cerebral arterial bed. It theoretically describes time needed for arterial blood after each heart stroke to arrive at hypothetical arteriole– capillary border in brain. It may be expressed in seconds or in a fraction of heart cycle (average in volunteers is 20–30%10). Previously we demonstrated that time constant shortens in hypercapnia in volunteers10 and prolongs (experimentally) when CPP decreases.9 In disease, it has been reported that time constant becomes shorter in carotid artery
stenotic disease17 and in cerebral vasospasm after subarachnoid haemorrhage.11 In the present study, we wanted to investigate potential links between CSF compensatory parameters and cerebral hemodynamic indices associated with t. We also wanted to investigate changes in cerebral hemodynamics during the infusion study. Previously, we described a relationship between cerebral autoregulation and resistance to CSF outflow.6 Patients with normal CSF circulation tend to have more frequently disturbed autoregulation, which probably reflects an existing cerebrovascular disease. However, time constant describes different properties of brain’s vasculature – associated with fast blood transport (fraction of seconds) rather than slow (seconds or even tens of seconds) regulation of vascular tone. Neither of indices (t, CVR, Ca)
Table 2 Comparison between baseline and plateau phase of infusion test (50 patients). Results are presented as median, lower quartile (Q1), and upper quartile (Q2) Median (Q1; Q3) baseline ICP [mmHg] ABP [mmHg] CPP [mmHg] CBFV [cm/s] CVR [mmHg/(cm/s)] Ca [cm/mmHg] t [s]
8.7 (2.6; 11.2) 89.9 (77.8; 103.3) 81.1 (72.6; 94.5) 48.3 (41.6; 62.4) 1.81 (1.28; 2.03) 0.11 (0.08; 0.16) 0.20 (0.14; 0.24)
Median (Q1; Q3) plateau 20.3 94.0 76.0 45.6 1.65 0.11 0.18
(16.6; 27.2) (84.3; 119.1) (63.0; 97.8) (37.25; 60.0) (1.34; 2.16) (0.82; 0.15) (0.14; 0.23)
Paired sign rank test 5.32610210 0.000003 0.0002 4.6361027 0.045 0.76 0.014
ICP: intracranial pressure; ABP: arterial blood pressure; CPP: cerebral perfusion pressure; CBFV: cerebral blood flow velocity; CVR: cerebrovascular resistance; Ca: cerebral arterial compliance (both CVR and Ca values are presented per unit of cross-sectional area of MCA [cm2]); t: cerebrovascular time constant.
258
Neurological Research
2014
VOL .
36
NO .
3
Capel et al.
Cerebrovascular time constant in hydrocephalus
Figure 2 Reciprocal character of percentage changes in cerebral arterial compliance (Ca) and cerebrovascular resistance (CVR) during infusion test. Variation in CVR (68%) prevails over changes in Ca (49%).
correlated with any of CSF compensatory parameters. Cerebral hemodynamic indices describe conditions to arterial blood inflow, CSF compensatory parameter conditions for CSF flow – which has a rate almost 2000 times lower than CBF. Our previous study18 indicated that Ca does not change during infusion test in contrast to decreasing cerebrospinal compartmental compliance. But cerebrovascular time constant has been never considered in hydrocephalus scenario. Infusion test modifies t, which gets significantly shorter. This shortening is a result of a significant decrease in CVR and non-significant increase in Ca. The decrease in t and increase in Ca are dependent on changes in CPP (Fig. 3A and C) but are independent of changes in ICP during the infusion test. The increase in ICP during infusion could not have been strong enough to activate changes in Ca, or/and the reactivity of cerebral arterial compliance to ICP changes might have been weakened by ongoing pathological process. In result, t is not correlated with changes in ICP within the investigated range of pressure variations. Ca and CVR usually change in opposite directions8 (see Fig. 2). During the infusion study, a decrease in CPP causes gradual vasodilatation (decrease in CVR) and an increase in cerebral arterial compliance (due to relaxation of arterial-arterioral smooth muscle). In experimental conditions,9 we found that an increase in Ca is deeper than a decrease in CVR; therefore t becomes longer. In hydrocephalus, a decrease in CVR prevails over an increase in Ca and t becomes shorter (swing of CVR is 68%, whereas Ca is only 49%). In experimental rabbits, the magnitude of a decrease in CPP during the infusion test was bigger than in patients with NPH. In fact, in 10 patients CPP increased during the infusion study, whereas such a situation did not have place in rabbits. Furthermore, rabbits were anesthetized, whereas
Figure 3 Relationships between changes induced by infusion in cerebral perfusion pressure (CPP [mmHg]) and percentage change in (A) cerebral arterial compliance (Ca [%]), (B) cerebrovascular resistance (CVR [%]), and (C) absolute changes in cerebral arterial time constant (t [s]). Decrease in CPP during infusion causes a decrease in CVR due to a vasodilatation of distal vessels and an increase in Ca. Changes in CVR prevail over changes in Ca; therefore, t varies in the same direction as CVR.
patients were conscious. Anesthetics may have influenced vessels’ tone and their reactivity. Finally, the reactivity of compliance may have been affected (more probably – at larger diameter vessels) by the process of hydrocephalus. It remains to be investigated whether inversed balance between changes in Ca and CVR in normal rabbits and hydrocephalic patients has anything in common with patophysiology of hydrocephalus.
Neurological Research
2014
VOL .
36
NO .
3
259
Capel et al.
Cerebrovascular time constant in hydrocephalus
Cerebral perfusion pressure and cerebral arterial time constant: the influence of vascular reactivity During the infusion test, CPP on average decreased even if 10 patients presented a slight increase in this parameter, due to a rise in ABP previously described as the early Cushing response.19 As a result of CPP decreasing, we observed a decreasing CVR. Conversely, with increasing CPP, CVR increased (see Fig. 3B). Decreasing of CVR indicates a vasodilatation of distal vessels allowing to maintain a constant CBF. This may indicate that vascular reactivity is in general preserved in NPH patients.
Arterial remodeling effect In a recent study,20 the authors found a relation between arterial stiffness and CVR, which confirms that increasing Ca is correlated with decreasing CVR. In another study,21 the authors showed that atherosclerosis and CVR could be affected by different etiologies, as activity of antioxidant enzymes. It could explain the inverse correlation between Ca and CVR. Arterial remodeling, arterial wall calcifications, and atherosclerosis can affect Ca and CVR, independently of CPP. Another mechanism can be explored after our preliminary work. Idiopathic orthostatic intolerance patients present a decreasing MCA flow velocity response to head-up tilt. This effect may be explained as impairment of cerebrovascular regulation.22 Cerebral arterial time constant depends physiologically on CPP and PaCO2 and in pathology – on many factors responsible or associated with developing disease. Cerebral arterial time constant is not dependant on cross-sectional area of insonated vessel and also independent of angle of insonation of the vessel. t derivates from the time constant of electric resistor–capacitor circuit.9–11 It is the interval required for a circuit or physiological system to switch from one state to another.
Limitations Calculations on cerebral arterial compliance (Ca) assumed that the venous outflow is constant during a cardiac cycle. Phase-contrast MRI study23 showed pulsatility of venous outflow in superior sagittal sinus (SSS) and jugular vein. Pulsatility increased from SSS to jugular vein. This pulsatility is lower than arterial inflow. Cross-sectional area of the insonated large vessel is not constant. It consistently changed Ca and CVR, but t is independent of this area. Finally, ABP was recorded from a peripheral artery with the pulsatile blood pressure (fundamental harmonic). It is not an ideal approximation of cerebral ABP.24 Our thesis that cerebral arterial time constant changes during the infusion study because of predominantly intact cerebrovascular reactivity, i.e.
260
Neurological Research
2014
VOL .
36
NO .
3
positive association between CVR and CPP, does not exhaust all questions about possible links between the nature of the disease and regulation of blood flow. We previously demonstrated possible disturbance of autoregulation in white matter.25 There is documented association between autoregulation and hormonal activity26 and autonomic drive.27 It is possible that the results of the study are interfered by inaccuracy of method of the cerebral arterial time constant estimation, and a link between compensatory parameters and cerebral hemodynamic indices might exist. Currently, TCD ultrasonography is the only method that allows for continuous measurement of changes in cerebral hemodynamics during the infusion study. Imaging techniques, although more accurate, can provide only a snapshot of the state of cerebral hemodynamics and their use during the infusion study is difficult due to technical limitations.
Conclusion The cerebrovascular time constant changes during the infusion study mainly due to autoregulatory CVR response to changing CPP. t is not dependent on CSF dynamics in NPH.
Conflict of Interest ICMz is a clinical neuroscience software supporting brain monitoring. It is licensed by University of Cambridge, Cambridge Enterprise Ltd (www.neurosurg.cam.ac.uk/icmplus). PS and MC have a financial interest in a fraction of licensing fee.
Acknowledgements Many thanks to Drs Christina Haubrich and Ruwan Weerakkody for assisting in collection of clinical data and all colleagues working for Hydrocephalus Clinic for their help, friendly advice, and clinical data collection. MK was supported by the Polish Ministry of Science and Higher Education.
References 1 Hakim S, Adams RD. The special clinical problem of symptomatic hydrocephalus with normal cerebrospinal fluid pressure: observations on cerebrospinal fluid hydrodynamics. J Neurol Sci. 1965;2:307–27. 2 Borgesen SE, Gjerris F. The predictive value of conductance to outflow of CSF in normal pressure hydrocephalus. Brain. 1982;105:65–86. 3 Boon AJ, Tans JT, Delwel EJ, Egeler-Peerdeman SM, Hanlo PW, Wurzer HA, et al. Dutch normal-pressure hydrocephalus study: the role of cerebrovascular disease. J Neurosurg. 1999;90:221–6. 4 Eide PK, Brean A. Cerebrospinal fluid pulse pressure amplitude during lumbar infusion in idiopathic normal pressure hydrocephalus can predict response to shunting. Cerebrospinal Fluid Res. 2010;7:5. 5 Kristensen B, Malm J, Fagerlund M, Hietala SO, Johanson B, Ekstedt J, et al. Regional cerebral blood flow, white matter abnormalities, and cerebrospinal fluid hydrodynamics in patients with idiopathic adult hydrocephalus syndrome. J Neurol Neurosurg Psychiatry. 1996;60:282–8.
Capel et al.
6 Czosnyka Z, Czosnyka M, Whitfield P, Donovan T, Pickard J. Cerebral autoregulation among patients with symptoms of hydrocephalus. Neurosurgery. 2002;50:526–33. 7 Czosnyka M, Richards HK, Reinhard M, Steiner LA, Budohoski K, Smielewski P, et al. Cerebrovascular time constant: dependence on cerebral perfusion pressure and endtidal carbon dioxide concentration. Neurol Res. 2012;34(1):17– 24. 8 Kim DJ, Kasprowicz M, Carrera E, Castellani G, Zweifel C, Lavinio A, et al. The monitoring of relative changes in compartmental compliances of brain. Physiol. Meas. 2009;30:647–59. 9 Carrera E, Kim DJ, Castellani G, Zweifel C, Smielewski P, Pickard JD, et al. Effect of hyper- and hypocapnia on cerebral arterial compliance in normal subjects. J Neuroimaging. 2011;21:121–5. 10 Kasprowicz M, Diedler J, Reinhard M, Carrera E, Steiner LA, Smielewski P, et al. Time constant of the cerebral arterial bed in normal subjects. Ultrasound Med Biol. 2012;38(7):1129–37. 11 Kasprowicz M, Czosnyka M, Soehle M, Smielewski P, Kirkpatrick PJ, Pickard JD, et al. Vasospasm shortens cerebral arterial time constant. Neurocrit Care. 2012 Apr;16(2):213–8. 12 Børgesen SE, Albeck MJ, Gjerris F, Czosnyka M, Laniewski P. Computerized infusion test compared to steady pressure constant infusion test in measurement of resistance to CSF outflow. Acta Neurochir (Wien). 1992;119:12–6. 13 Czosnyka M, Pickard JD. Monitoring and interpretation of intracranial pressure. J Neurol Neurosurg Psychiatry. 2004;75:813–21. 14 Czosnyka M, Wollk-Laniewski P, Batorski L, Zaworski W. Analysis of intracranial pressure waveform during infusion test. Acta Neurochir (Wien). 1988;93:140–5. 15 Weerakkody RA, Czosnyka M, Schuhmann MU, Schmidt E, Keong N, Santarius T, et al. Clinical assessment of cerebrospinal fluid dynamics in hydrocephalus. Guide to interpretation based on observational study. Acta Neurol Scand. 2011;124(2):85–98. 16 Smielewski P, Lavinio A, Timofeev I, Radolovich D, Perkes I, Pickard JD, et al. ICMz, a flexible platform for investigations of cerebrospinal dynamics in clinical practice. Acta Neurochir Suppl. 2008;102:145–51. 17 Kasprowicz M, Diedler J, Reinhard M, Carrera E, Smielewski P, Budohoski KP, et al. Time constant of the cerebral arterial bed. Acta Neurochir Suppl. 2012;114:17–21.
Cerebrovascular time constant in hydrocephalus
18 Kim DJ, Carrera E, Czosnyka M, Keong N, Smielewski P, Bale´dent O, et al. Cerebrospinal compensation of pulsating cerebral blood volume in hydrocephalus. Neurol Res. 2010;32(6):587–92. 19 Schmidt EA, Czosnyka Z, Momjian S, Czosnyka M, Bech RA, Pickard JD. Intracranial baroreflex yielding an early Cushing response in human. Acta Neurochir Suppl. 2005;95:253–6. 20 Robertson AD, Tessmer CF, Hughson RL. Association between arterial stiffness and cerebrovascular resistance in the elderly. J Hum Hypertens. 2010;24:190–6. 21 D’Armiento FP, Bianchi A, de Nigris F, Capuzzi DM, D’Armiento MR, Crimi G, et al. Age-related effects on atherogenesis and Scavenger enzymes of intracranial and extracranial arteries in men without classic risk factors for atherosclerosis. Stroke. 2001;32:2472–80. 22 Jacob G, Atkinson D, Jordan J, Shannon JR, Furlan R, Black BK, et al. Effects of standing on cerebrovascular resistance in patient with idiopathic orthostatic intolerance. Am J Med. 1999;106:59–64. 23 Stoquart-ElSankari S, Lehmann P, Villette A, Czosnyka M, Meyer M-E, Deramond H, et al. A phase-contrast MRI study of physiologic cerebral venous flow. J Cereb Blood Flow Metab. 2009;29:1208–15. 24 O’Rourke MF, Adji A. Noninvasive studies of central aortic pressure. Curr Hypertens Rep. 2012;14(1):8–20. 25 Momjian S, Owler BK, Czosnyka Z, Czosnyka M, Pena A, Pickard JD. Pattern of white matter regional cerebral blood flow and autoregulation in normal pressure hydrocephalus. Brain. 2004;127(5):965–72. 26 Hong KW, Shin HK, Kim CD, Lee WS, Rhim BY. Restoration of vasodilation and CBF autoregulation by genistein in rat pial artery after brain injury. Am J Physiol Heart Circ Physiol. 2001;281(1):308–15. 27 Lavinio A, Ene-Iordache B, Nodari I, Girardini A, Cagnazzi E, Rasulo F, et al. Cerebrovascular reactivity and autonomic drive following traumatic brain injury. Acta Neurochir Suppl. 2008;102:3–7. 28 Wikkelsø C, Hellstro¨m P, Klinge PM, Tans JT; European iNPH Multicentre Study Group. The European iNPH Multicentre Study on the predictive values of resistance to CSF outflow and the CSF Tap Test in patients with idiopathic normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry. 2013 May;84(5):562–8.
Neurological Research
2014
VOL .
36
NO .
3
261
