Cine-PC MR in assessment of cerebrospinal fluid velocity in the aqueduct of the midbrain correlated with intracranial pressure – Initial study

Cine-PC MR in assessment of cerebrospinal fluid velocity in the aqueduct of the midbrain correlated with intracranial pressure – Initial study

Medical Hypotheses 78 (2012) 227–230 Contents lists available at SciVerse ScienceDirect Medical Hypotheses journal homepage: www.elsevier.com/locate...

337KB Sizes 0 Downloads 8 Views

Medical Hypotheses 78 (2012) 227–230

Contents lists available at SciVerse ScienceDirect

Medical Hypotheses journal homepage: www.elsevier.com/locate/mehy

Cine-PC MR in assessment of cerebrospinal fluid velocity in the aqueduct of the midbrain correlated with intracranial pressure – Initial study BO Zhang, Song-bai Li ⇑ Department of Radiology, The First Affiliated Hospital of China Medical University, Shenyang, Liaoning 110001, PR China

a r t i c l e

i n f o

Article history: Received 10 August 2011 Accepted 23 October 2011

a b s t r a c t We assessed the changes in cerebrospinal fluid (CSF) hydrodynamics caused by barriers to CSF circulation, and determined the relationship between CSF velocity and intracranial pressure in the aqueduct of the midbrain. This was determined by correlating the CSF peak flow velocity with the intracranial pressure (ICP) obtained from a lumbar puncture (LP) procedure. The CSF peak flow velocity was measured by finger pulse-gated cine-phase contrast (PC) MR scan 8–12 hours after LP was performed in 28 patients. All patients were divided into 2 groups based on the directional patterns of the CSF net flow in the aqueduct of the midbrain over one cardiac cycle. The CSF peak net velocity (Vnet) was then correlated with ICP utilizing Pearson correlation analysis method, with significance difference assigned at the 5% level. Routine MR scanning revealed no abnormal findings in the brain when the direction of the CSF net flow is caudal. Vnet in the aqueduct of the midbrain was correlated positively with ICP (y(V) = 0.011 + 0.002(ICP), r = 0.69, p < 0.01). However, varying degrees of the hydrocephalus were observed in those patients who demonstrated a cranial direction of the CSF net flow. Our results indicate that non-invasive measurement of the CSF peak flow with cine-PC MR imaging can be related to the change of CSF circulation caused by the obstructions to the CSF circulation in the patients with various neurological disorders. This unique method may be a substantially useful tool to assess the changes in the ICP in the directional pattern. Ó 2011 Elsevier Ltd. All rights reserved.

Introduction As one of common clinical procedures, lumbar puncture (LP) has been widely used and yielded very useful information which can be used to interpret physiological, biochemical and pathological profiles of the cerebrospinal fluid (CSF) including intracranial pressure (ICP) and the contents of CSF. However, due to the invasive nature of LPs, the procedure may not be always accepted by the patients and their family members. The procedure itself requires strict aseptic manipulation, and special care must be given to avoid microvascular or spinal cord damage during large pore needle penetration. Furthermore, it is an absolute contraindication to those patients who are presenting with shock, skin infections located in the puncture site, and to patients who have a significant rise in intracranial pressure (ICP) while exhibiting clinical signs of herniation. Although LP is a valuable tool in assessing the characteristics of CSF in various conditions, little information regarding CSF hydrodynamics, particularly CSF peak flow velocity in the subregions of intracranial space of the CSF system, can be provided. Since its first use in the evaluation of cardiovascular circulation [1–4], cine phase-contrast (PC) MR imaging has been implemented for the assessment of CSF flow in a variety of normal and abnormal ⇑ Corresponding author. Tel.: +86 13478180253. E-mail address: [email protected] (S.-b. Li). 0306-9877/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.mehy.2011.10.031

conditions, and has shown relatively high accuracy and reproducibility for measurements of CSF flow [5–7]. This method offers a great advantage over LP due to its non-invasive functional imaging and allows the measurement of CSF hydrodynamics in many patients including those clinically contradicted to LP procedure. However, it is still unknown if the changes in CSF peak flow velocity correlate with changes in ICP in various disorders caused by obstructions or barriers to CSF circulation. We therefore hypothesized that there is a relationship between (ICP) and CSF dynamics, and that such a correlation could be affected by changes in CSF dynamics following any disturbance to CSF circulation. In the present study, we validated our hypotheses by investigating the changes in the hydrodynamics of CSF flow in a total of 28 patients with a variety of neurological disorders by means of cine-PC MR technique and by assessing the relationship between CSF peak flow velocity in the aqueduct of the midbrain and ICP obtained from LP. Methods Subjects From October 2010 to January 2011, cine-PC MRI was performed on a total of 28 patients admitted to our hospital with a variety of neurological disorders. The study population consisted

228

B. Zhang, S.-b. Li / Medical Hypotheses 78 (2012) 227–230

of 14 male and 14 female patients aged 16–70 (mean: 47.0 years). The arterial blood pressures measured prior to the beginning of MRI imaging procedures were under 150/90 mmHg and heart rates were between 60 and 90 b/m. Since the CSF pulsates within the craniospinal axis in response to rhythmic alterations of cerebral blood circulation during the cardiac cycle, patients with arrhythmia were not included in this study. The CSF flow was measured within 8–12 h post LP procedure and no patients received treatment for the management of intracranial pressure prior to the measurement of ICP and the CSF flow. Only 5 ml of CSF was withdrawn from each patient and this small volume loss of CSF could not potentially impact the subsequent measurement of the CSF flow as CSF regenerates every 6–8 h. Informed consents were obtained from all the patients or their first degree family members or relatives before the beginning of the study. Clinical diagnosis Demographic and clinical data of all 28 patients involved in this study are summarized in Table 1. Among these neurological disorders, encephalitis or myelitis, cerebrovascular accidents and peripheral neuropathy were frequently encountered.

(TE), minimum; field of view (FOV), 22  22 cm; matrix, 256  256 and flip angle, 20°. The phase encoding direction was set from cranial to caudal. The peak velocity, flow, direction and duration of CSF circulation were analyzed and calculated with the ReportCARD software. In the velocity-coded images, cranial-directed flow showed low signal intensity with a negative velocity (Fig. 1A); whereas the caudal-directed flow had high signal intensity with a positive velocity (Fig. 1B). The CSF circulation time was defined as the interval between the two start time points of the caudal-directed CSF flow. Mean peak velocity in a cardiac systolic phase, Vs, and mean peak velocity in a cardiac diastolic phase, Vd, were computed from 30 time points of the caudal-(Vs1–Vs30) and cranial-directed (Vd1–Vd30) CSF flow, respectively. The CSF peak net velocity (Vnet = |Vs| |Vd|) during a single cardiac cycle, the caudal-Fs (CSF flow during a cardiac systolic phase) and cranial-directed-Fd (CSF flow during a cardiac diastolic phase) were all obtained based on ReportCard software, respectively. (Fig. 2/Table 2). Accordingly, all patients involved in this study were divided into 2 groups based on the direction of CSF net flow: In group 1, there were 22 patients with net flow of greater than zero (Fnet = |Fs| |Fd| > 0) in the aqueduct of the midbrain during a single cardiac cycle. The remaining six patients with a net flow of less than zero (Fnet = |Fs| |Fd| < 0) were included in group 2. Statistical analysis

Design Using a peripheral pulse transducer applied to the subject’s finger, CSF peak flow rate was measured simultaneously with a Signa HDxt 3.0T superconductive magnetic resonance machine (GE, America). Thirty time-point images were collected throughout each cardiac cycle to create 30 velocity-coded images from the region of interest (ROI) covering the aqueduct of the midbrain (Fig. 1). The parameters used for data collection were as follows: velocity encode, 15 cm/s; repeat time (TR), minimum; echo time

Table 1 Summary of patient data. Case No.

Age (years), sex

Clinical diagnosis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

57, 23, 45, 42, 70, 48, 49, 16, 52, 31, 39, 59, 37, 59, 59, 64, 28, 49, 45, 47, 24,

F F M F F M M M M M F F F F F F M F M M F

22 23 24 25 26 27

50, 56, 57, 75, 56, 48,

F M M F M M

28

69, M

Encephalitis Encephalitis Encephalitis Myelitis Myelitis Myelitis Myelitis Myelitis Ischemic cerebrovascular disease Ischemic cerebrovascular disease Ischemic cerebrovascular disease Ischemic cerebrovascular disease Ischemic cerebrovascular disease Ischemic cerebrovascular disease Peripheral neuropathy Peripheral neuropathy Peripheral neuropathy Peripheral neuropathy Peripheral neuropathy Motor neuron disease Superficial siderosis in the central nervous system Intracranial hypotension syndrome Purulent meningitis with hemorrhagic infarction Purulent meningitis with hemorrhagic infarction Transsphenoidal surgery of pituitary fossa tumors Transsphenoidal surgery of pituitary fossa tumors Ischemic cerebrovascular disease with arachnoid cyst Peripheral neuropathy with hydrocephalus

Data were analyzed using Pearson correlation analysis of SPSS 17.0, with significant difference between groups assigned at the 5% level. Results We completed our study protocols and obtained finger-pulse gated PC-MR imaging data for all 28 patients with a variety of neurological disorders. The gender ratio of the patient population was 1. ICP from most patients (n = 24) was within normal range (80 mmH2O–180 mmH20). The remaining patients had ICP of 70 mmH2O (n = 2), 75 mmH2O (n = 1) and 225 mmH2O (n = 1), respectively. In the patients in group 1 (n = 22), the normal images of the brain were detected from routine MR scanning when the direction of CSF net flow is craniocaudal (n = 22: Fnet > 0). In contrast, mild to severe hydrocephalus was observed in group 2 (n = 6: Fnet < 0). Furthermore, we found that there was a clear relationship between the CFS peak velocity and the ICP, and this relationship is associated with the directional patterns of CSF flow as demonstrated in Fig. 3. For example, Vnet in the aqueduct of the midbrain was correlated positively with ICP (Fig. 3: y(Vnet) = 0.011 + 0.002(ICP), r = 0.69, p < 0.01) when the direction of the CSF net flow was craniocaudal, occurring with no significant or abnormal findings during the routine MR scanning (Fig. 3). However, we failed to detect a correlation between Vnet and ICP when the direction of the CSF net flow was cranial (n = 6) (r = 0.318, p = 0.682). The diagnosis of these 6 patients included purulent meningitis with hemorrhagic infarction (n = 2), transsphenoidal surgery of pituitary fossa tumors (n = 2), ischemic cerebrovascular disease with arachnoid cyst (n = 1) and peripheral neuropathy with hydrocephalus (n = 1). These pathological profiles suggest that the formation of barriers or obstructions in the CSF canal may have contributed to the changes observed in the CSF circulation. . Discussion and limitations Cine-PC MR has been proven to be an effective method in the evaluation of CSF flow in humans [8]. In our study we chose the plane of the aqueduct of the midbrain which is relatively regular

B. Zhang, S.-b. Li / Medical Hypotheses 78 (2012) 227–230

229

Fig. 1. The velocity-coded images collected by axial phase-contrast MRI from a 45-year old male patient with peripheral neuropathy. 1A: A cranial-directed CSF flow in a cardiac diastolic phase, showing low signal intensity (arrow). 1B: A caudal-directed CSF flow in a systolic phase, showing high signal intensity (arrow).

Fig. 2. Graph shows the relationship between the CSF peak velocity (Vs/Vd) and the time throughout one cardiac cycle in the aqueduct of the midbrain obtained from cine-PC MR from a 64-year old female patient with peripheral neuropathy: the choice of retrospective gating provides up to 30 time-points equally distributed throughout the cardiac cycle.

Table 2 Data collected from the ROI at the aqueduct of the midbrain from the same patient.

Caudal-Vs (mm/s) Cranial-Vd (mm/s) Caudal-Fs (ml/b) Cranial-Fs (ml/b)

V1

V2

V3

V4

V5

...

V28

V29

V30

13 0 0.023 0.019

13 0

18 0

22 0

25 0

... ...

0

0

13 0

17.8

11.6

Vs and Vd was calculated as follows, respectively: Vs = (Vs1 + Vs2 + Vs3 + . . . + Vs29 + Vs30)/30. Vd = (Vd1 + Vd2 + Vd3 + . . . + Vd29 + Vd30)/30.

in anatomical shape and a law of the CSF through the aqueduct of the midbrain has been found [9]. Previous studies [10,11,1] have confirmed that CSF flows from the cranial cavity into the vertebral canal during the cardiac systolic phase because of dilation of the brain, and then moves back into the cranial cavity during the diastolic phase owning to retraction of the brain. The arterial expansion creates the prerequisites for expansion of the brain by venting CSF to the spinal canal. The expansion of the brain, in turn, is responsible for compression of the ventricular system and hence the intraventricular flow of CSF. Thus, the CSF in the cranial cavity flows toward the spinal space in a back-and-forth manner in response to the pulsations of the brain. CSF dynamics throughout the aqueduct of the midbrain can be modulated by various factors [11]. The relatively important ones include cardiac rhythm, heart rate, blood pressure and intracranial

diseases. Heart rate within a relatively wide range (>100 b/m, <60 b/m) will affect the immediate velocity of CSF because of large variations in each cardiac cycle. In addition, dynamic parameter errors could be made by cine-PC when the data are collected during premature contraction of patients with arrhythmia. Furthermore, it has also been demonstrated that patients with systolic pressure over 150 mmHg present hyperdynamics of CSF flow [12]. Therefore, we only included the patients with blood pressure under 150/90 mmHg and heart rate range from 60 to 90b/m with normal heart rhythm in our study. In doing so, the interference of cardiac irregularity or hypertension on the measurement of CSF flow could be kept to the minimum. A large number of studies so far have focused on the role of cine-PC MRI imaging in making a diagnosis and developing surgical treatment plans for brain disorders, such as normal pressure hydrocephalus, [13,14] or a group of congenital and acquired disorders including Chiari malformation, syringomyelic cyst and arachnoid cyst [15]. All of these brain abnormalities can alter hydrodynamics of CSF circulation. In our present study, however, we found that hydrodynamics of CSF flow can be altered even in the absence of obvious intracranial space-occupying lesions. This is a very interesting observation and more studies are certainly warranted to explore it further. The most striking finding in the present study was the distinct pattern in the routine MR scanning between the two groups based on the direction of CSF net flow. The routine MR scanning shows

230

B. Zhang, S.-b. Li / Medical Hypotheses 78 (2012) 227–230

Fig. 3. Graph shows the relationship between Vnet in the aqueduct of the midbrain obtained from cine-PC MR and ICP based on LP data (n = 22). (y(Vnet) = 0.011 + 0.002(ICP), r = 0.69, p < 0.01).

varying degrees of hydrocephalus in the 6 patients in our study when the direction of CSF net flow is cranial, whereas no abnormality was identified with the routine MRI scanning when the direction of the CSF flow is caudal. The results indicate that the barriers or obstructions to CSF circulation may be formed due to arachnoid adhesions in the CSF pathway, in turn; the direction of CSF net flow over a cardiac cycle was altered. However, it should be noted that data obtained from another study [6] suggests that the normal CSF net flow in a cardiac cycle could be caudal as well. On the other hand, previously published data [16,17] have suggested that it is not uncommon to have a cranial direction of the intracranial CSF net flow over a cardiac cycle in patients with neurological disorders such as communicating hydrocephalus or normal pressure hydrocephalus. The results from our current study are consistent with the above findings and support our hypothesis. Another finding of the present study is that there is a relationship between Vnet in the aqueduct of the midbrain and actual ICP, and the net flow value depends on the directional pattern of the CSF flow. The Vnet is correlated positively with ICP when the direction of CSF net flow is caudal and no significant abnormalities are detected from the routine MR scanning. This finding suggests that the interrelationship between intracranial pressure and CSF flow in the brain could serve as an important interface, through which the hydrodynamics of CSF circulation can be related to the changes in ICP. Conclusion Our study demonstrates a correlation of the CSF peak velocity with intracranial pressure (ICP) in patients with some neurological disorders. Non-invasive measurements of CSF hydrodynamics by cine-PC MR imaging make this technique more suitable for clinical practice. Acknowledgment We gratefully acknowledge Dr. Qiang Li, MD., PhD from Duke University Medical Center for his valuable comments on the manuscript, and Dr. Louise Risher, PhD and Mr. Trevor Johnson, MS from Duke University Medical Center for their assistance in preparing the manuscript. The authors also thank those patients at China Medical University whose participation made this study possible.

References [1] Enzmann DR, Ross MR, Marks MP, Pelc NJ. Blood flow in major cerebral arteries measured by phase-contrast cine MR. AJNR Am J Neuroradiol 1994;15:123–9. [2] Feinberg DA, Crooks L, Hoenninger 3rd J, Arakawa M, Watts J. Pulsatile blood velocity in human arteries displayed by magnetic resonance imaging. Radiology 1984;153(1):177–80. [3] Meier D, Maier S, Bosiger P. Quantitative flow measurements on phantoms and on blood vessels with MR. Magn Reson Med 1988;8(1):25–34. [4] Firmin DN, Nayler GL, Klipstein RH, Underwood SR, Rees RS, Longmore DB. In vivo validation of MR velocity imaging. J Comput Assist Tomogr 1987;11(5): 751–6. [5] Wentland AL, Wieben O, Korosec FR, Haughton VM. Accuracy and reproducibility of phase-contrast MR imaging measurements for CSF flow. AJNR Am J Neuroradiol 2010;31:1331–6. [6] Enzmann DR, Pelc NJ. Normal flow patterns of intracranial and spinal cerebrospinal fluid defined with phase-contrast cine MR imaging. Radiology 1991;178(2):467–74. [7] Quencer RM, Post MJ, Hinks RS. Cine MR in the evaluation of normal and abnormal CSF flow: intracranial and intraspinal studies. Neuroradiology 1990;32(5):371–91. [8] Zhu DC, Xenos M, Linninger AA, Penn RD. Dynamics of lateral ventricle and cerebrospinal fluid in normal and hydrocephalic brains. J Magn Imaging 2006;24(4):756–70. [9] Sweetman B, Linninger AA. Cerebrospinal fluid flow dynamics in the central nervous system. Ann Biomed Eng 2010;39(1):484–96. [10] Ohara S, Nagai H, Matsumoto T, Banno T. MR imaging of CSF pulsatory flow and its relation to intracranial pressure. J Neurosurg 1988;69(5):675–82. [11] Greitz D, Wirestam R, Franck A, Nordell B, Thomsen C, Stahlberg F. Pulsatile brain movement and associated hydrodynamics studied by magnetic resonance phase imaging. The Monro-Kellie doctrine revisited. Neuroradiology 1992;34(5):370–80. [12] Yao CC, Liang CS, Wang YJ. CSF flow study by phase-contrast cine MRI. Nerv Dis Ment Health 2002;1(2):57–9. [13] Sharma AK, Gaikwad S, Gupta V, Garg A, Mishra NK. Measurement of peak CSF flow velocity at cerebral aqueduct, before and after lumbar CSF drainage, by use of phase-contrast MRI: utility in the management of idiopathic normal pressure hydrocephalus. Clin Neurol Neurosurg 2008;110(4):363–8. [14] Mounton Paradot G, Baledent O, Sallioux G, Lehmann P, Gondry-Jouet C, Le Gars D. Contribution of phase-contrast MRI to the management of patients with normal pressure hydrocephalus: can it predict response to shunting? Neurochiruqie 2010;56(1):50–4. [15] Battal B, Kocaoglu M, Bulakbasi N, Husmen G, Tuba Sanal H, Tayfun C. Cerebrospinal fluid flow imaging by using phase-contrast MR technique. Br J Radiol 2011;84(1004):758–65. [16] Yao WW, Xingrong Chen, Tianzhen Shen. MRI quantitative estimation of CSF in hydrocephalus. J Clin Radiol 2002;21(10):12–6. [17] Penn RD, Basati S, Sweetman B, Guo X, Linninger A. Ventricle wall movements and cerebrospinal fluid flow in hydrocephalus. J Neurosurg 2011;115(1): 159–64.