CLINICAL STUDIES

                                     EVALUATING THE EFFECT OF DECOMPRESSION SURGERY
                                     ON CEREBROSPINAL FLUID FLOW AND INTRACRANIAL
                                     COMPLIANCE IN PATIENTS WITH CHIARI MALFORMATION
                                     WITH MAGNETIC RESONANCE IMAGING FLOW STUDIES

Anusha Sivaramakrishnan,              OBJECTIVE: To quantify the effect of decompression surgery on craniocervical junc-
M.S.                                  tion hydrodynamics and on global intracranial compliance (ICC) in patients with
Department of Radiology,              Chiari I malformation by use of magnetic resonance measurements of cerebrospinal
University of Illinois at Chicago,
Chicago, Illinois
                                      fluid and blood flow. Studying the effect of decompression surgery may improve our
                                      understanding of the pathophysiological characteristics of Chiari I malformation and
Noam Alperin, Ph.D.                   aid in identifying patients who will benefit from the procedure.
Department of Radiology,              METHODS: Twelve patients were studied with a 1.5-T magnetic resonance imaging
University of Illinois at Chicago,    scanner before and after decompression surgery. Cine phase contrast magnetic reso-
Chicago, Illinois
                                      nance images were used to quantify maximum cord displacement, maximum systolic
Sushma Surapaneni, B.S.               cerebrospinal fluid velocity and volumetric flow rate, and overall ICC. ICC was derived
Department of Radiology,
                                      by use of a previously reported method that measures small changes in intracranial
University of Illinois at Chicago,    volume and pressure that occur naturally with each cardiac cycle.
Chicago, Illinois
                                      RESULTS: After surgery, changes were documented both in the local hydrodynamic
                                      parameters and in ICC. However, only the change in ICC, an average increase of more
Terry Lichtor, M.D., Ph.D.
                                      than 60%, was statistically significant. Increased ICC, which was associated with
Department of Neurosurgery,
Rush-Presbyterian St. Luke’s          improved outcome, was measured in 10 of the 12 patients, no significant change was
Medical Center,                       documented in 1 patient, and decreased ICC was measured in 1 patient whose
Cook County Hospital,
                                      symptoms persisted after surgery.
Chicago, Illinois
                                      CONCLUSION: An increase in the overall compliance of the intracranial compart-
Reprint requests:                     ment is the most significant and consistent change measured after decompression
Noam Alperin, Ph.D.,
                                      surgery. Changes in cord displacement, cerebrospinal fluid velocities, and flow in the
Physiological Imaging and
Modeling Laboratory, Department       craniospinal junction were less consistent and less affected by the operation. Thus, ICC
of Radiology (M/C 711), University    may play an important role in the outcome of decompression surgery related to
of Illinois at Chicago,
                                      improving symptoms and restoring normal neurological hydrodynamics in patients
830 South Wood Street,
Chicago, IL 60612.                    with Chiari I malformations.
Email: alperin@uic.edu
                                      KEY WORDS: Cerebrospinal fluid flow, Chiari malformation, Cine phase contrast magnetic resonance
                                      imaging, Decompression surgery, Intracranial compliance
Received, February 27, 2004.
Accepted, August 19, 2004.            Neurosurgery 55:1344-1351, 2004   DOI: 10.1227/01.NEU.0000143612.60114.2D   www.neurosurgery-online.com

        hiari I malformation is characterized by displacement of         or syringomyelia, is more common. Typical symptoms in pa-
        the cerebellar tonsils more than 5 mm caudally through           tients with Chiari I malformation include headache, neck pain,
        the foramen magnum (8, 15). The level of hindbrain her-          and sensory and motor deficits. However, as no direct correlation
niation is best revealed by midline sagittal T1-weighted magnetic        has been observed between symptoms and the anatomic severity
resonance imaging (MRI), which currently is used for diagnosis           of herniation, the pathophysiological characteristics of the dis-
of the disorder (13). Chiari I malformation is not commonly              ease cannot be explained by the abnormal anatomy of the cra-
associated with other cerebral anatomic abnormalities, although          niospinal junction alone.
association with hydrocephalus has been reported (14). Associa-             Several investigators attempted to better understand the
tion with spinal canal abnormalities, such as progressive scoliosis      pathophysiological characteristics of Chiari I malformation by

1344 | VOLUME 55 | NUMBER 6 | DECEMBER 2004                                                                         www.neurosurgery-online.com
                                                 CHANGES     IN INTRACRANIAL    COMPLIANCE       WITH   DECOMPRESSION SURGERY

measuring cerebrospinal fluid (CSF) flow velocities at differ-       of the CSF spaces and the posture of the neck and global
ent locations around the craniospinal junction with cine phase       factors such as cerebral hemodynamics and the biomechanical
contrast MRI (7, 12, 18, 20, 22, 24). A quantitative study per-      state of the craniospinal system (2, 5, 11, 16, 23). Therefore,
formed by Shibuya et al. (24) in 31 patients with Chiari mal-        there is a clear need for further study of the functional effect of
formation demonstrated that the amplitude of CSF velocities          decompression surgery to analyze parameters that may be less
at the C3 level posterior to the cord decreases with increased       variable.
hindbrain herniation. Pujol et al. (22) reported higher tonsillar       Our group recently developed a system-based analysis of
pulsations with delayed upward CSF flow in patients with             MRI measurements of CSF and blood flow to characterize the
Chiari malformation compared with normal subjects. In a              craniospinal hydrodynamics as well as overall intracranial
similar study, Hofmann et al. (18) demonstrated an increase in       compliance (ICC). Decompression-related changes in the up-
maximum systolic and diastolic cord displacement rates and           per cervical spine hydrodynamics and in the overall ICC are
an impairment of upward flow in the anterior and posterior           measured and evaluated for their significance. The ICC is
CSF spaces of the spinal canal in patients with Chiari malfor-       derived from volumetric flow rates (VFRs) of blood and CSF
mations. Altered CSF flow velocities and cord motion dynam-          flow to and from the intracranial vault. The ICC measurement
ics are associated with Chiari I malformation; however, the          is based on the phasic changes in intracranial volume and
clinical relevance of these findings and their contribution to       pressure that occur naturally during the cardiac cycle. The
the understanding of the pathophysiological characteristics of       decompression surgery-related changes in ICC, maximum
the disorder has been limited.                                       value of mean CSF velocity at C2, maximum CSF VFR, CSF
   Although the disease process is not well understood, de-          volume that oscillates between the cranium and spinal canal,
compression surgery is now commonly offered as a treatment           and maximal cord displacement were compared and evalu-
of choice for patients with Chiari malformation with or with-        ated for their statistical significance.
out syringomyelia. Surgical options include decompression
with or without opening of the dura mater, opening of the                         PATIENTS AND METHODS
dura with or without closing or patching, lysis of adhesions
and exploration of the 4th ventricle with or without obex            Patients
plugging or stenting of the ventricle, resection of the tonsils,
and various shunting procedures for the syrinx itself (21). The         Twelve patients were studied before and after decompres-
outcome of decompression surgery usually is evaluated by             sion surgery (eight women and four men; age range, 30–58 yr;
comparing symptoms before and after surgery. Good outcome            mean age, 42       8 yr) by use of a 1.5-T MRI Signa scanner
is usually reported after surgery (6), although failed decom-        (General Electric Medical systems, Milwaukee, WI). The study
pression with either persistence or reoccurrence of symptoms         protocol was approved by the institutional review board, and
several months after decompression also has been reported            consent was obtained from all patients. Of these 12 patients, 4
(10, 17, 19).                                                        had Chiari I malformation alone, 5 had Chiari I malformation
   Badie et al. (9) analyzed the anatomic posterior fossa ratio as   and syringomyelia, and 3 had Chiari I malformation and
a means to predict response to suboccipital decompressive            hydrocephalus. One patient with syringomyelia had a history
surgery. They reported a smaller posterior fossa ratio in pa-        of craniofacial surgery for Crouzon’s disease. Symptoms in
tients with Chiari malformation compared with control sub-           these patients included headache, neck pain, and sensory and
jects and further demonstrated that a smaller posterior fossa        motor difficulties (such as vertigo, ataxia, and dysmetria). The
ratio is associated with better surgical outcome. Bhadelia et al.    tonsillar herniation ranged from 5 to 17 mm (mean, 9.3 mm;
(12) investigated the functional effect of decompression sur-        standard deviation, 3.8 mm). All patients underwent suboc-
gery by measuring systolic and diastolic CSF flow velocities at      cipital craniectomy, C1 laminectomy, and duraplasty. The
four different regions in the craniospinal junction and the          three patients with hydrocephalous did not undergo any ad-
upper cervical subarachnoid space with MRI . They reported           ditional procedures such as placement of a shunt, with the
a statistically significant increase in maximum CSF velocities       exception of one patient who had a temporary external ven-
after decompression surgery only in the anterior subarachnoid        tricular drainage placed during the operation, which was re-
space below the foramen magnum. Armonda et al. (7), how-             moved on postoperative Day 13. Eleven of the patients re-
ever, measured significant change in CSF velocity only dorsal        ported improvement in symptoms, and one patient (with
to the spinal cord at the level of C2. This inconsistency is not     syringomyelia) reported persistence of symptoms and weak-
surprising, as CSF flow velocity changes considerably from           ness and numbness in one hand.
one region to another within the complex anatomy of the
craniospinal junction.                                               MRI Parameters
   CSF velocity has been demonstrated to vary considerably             Standard anatomic sagittal T1-weighted MRI scans were
among individuals, even when measured in the same location           used to estimate herniation levels in all patients before sur-
(11, 25). This known intersubject variability in CSF velocity        gery. Hydrodynamic parameters and ICC were obtained from
may be attributable to the many factors that modulate CSF            measurements of CSF and blood flows to and from the cra-
flow. These include local factors such as the variable geometry      nium with low- and a high-velocity encoded cine phase con-

NEUROSURGERY                                                                               VOLUME 55 | NUMBER 6 | DECEMBER 2004 | 1345

trast scans, respectively. Blood flow was imaged by velocity
encoding ranging from 60 to 90 cm/s, and the slower CSF flow
was imaged by velocity encoding ranging from 5 to 9 cm/s.
The CSF flow was measured at mid C2 level, where the
luminal cross sectional area is relatively unchanged, to obtain
reliable velocity and VFR measurements. Other imaging pa-
rameters were as follows: slice thickness, 5 to 6 mm; field of
view, 14 to 16 cm; repetition time, 17 to 22 milliseconds; echo
time, 6.5 to 9 milliseconds; and flip angle of 20 and 25 degrees
for the CSF and blood flow scans, respectively.

Derivation of Hydrodynamic Parameters
   CSF flow between the cranium and spinal canal was mea-          FIGURE 1. A compartmental model proposed by Alperin et al. (1) used
                                                                   for the derivation of the intracranial volume change. The model shows
sured with the low-velocity encoding images. CSF velocities
                                                                   arterial inflow, venous outflow to the cranium, and oscillating CSF flow
inside the cross sectional area of the CSF space were averaged
                                                                   between the cranium and spinal canal. During systole, arterial blood
to obtain the mean CSF velocity. The VFR then was obtained         inflow is greater than venous and CSF outflow, causing the small change
by multiplying the mean velocity by the CSF space cross            in intracranial volume.
sectional area. The cord displacement rates during the cardiac
cycle were derived by multiplying the mean cord velocity by        the Navier-Stokes equation (3). An ICC index (CI) then was
its cross sectional area. Cord displacement was then obtained      calculated from the ratio of these small changes in intracranial
by integration with respect to time during the cardiac cycle.      volume and pressure that occur naturally with each cardiac
For increased measurement accuracy and reproducibility, lu-        cycle (3, 23).
men boundaries were delineated with a recently developed
automated method for segmentation of lumens conducting
pulsatile flow, the pulsatility-based segmentation technique       Data Analysis
(1). The parameters analyzed were maximum cord displace-              The effect of decompression surgery was quantified by
ment, maximum value of mean CSF velocity during the sys-           comparing changes in the analyzed parameters before and
tolic phase, and maximum value of CSF VFR during the               after surgery. Student’s single-tailed paired t test was per-
systolic phase. The amount of CSF volume that flows from and       formed on the pre- and postdecompression parameters for all
to the intracranial vault (i.e., oscillatory CSF volume) was       12 patients and separately for the 9 patients without hydro-
obtained by integrating the absolute values of the CSF flow        cephalus by use of SYSTAT software, Version 8 (SYSTAT
waveform during the cardiac cycle and dividing the sum by          Software, Inc., Richmond, CA). A P value of 0.05 was con-
two. Intersubject average and standard deviation before and        sidered a statistically significant difference.
after surgery then were calculated.

Derivation of ICC                                                                               RESULTS
   The neurophysiological basis for the derivation of the intra-      Sagittal T1-weighted anatomic images from patients with
cranial volume change during the cardiac cycle was explained       and without hydrocephalous, before and after decompression
according to the simplified compartmental model of the cra-        surgery are shown in Figure 2. An example of an MRI scan
niospinal system (Fig. 1). The intracranial volume change was      used for quantitation of the blood flow to and from the brain
derived from the instantaneous difference among volumetric         is shown in Figure 3. A scout magnetic resonance angiogram
arterial inflow, venous outflow, and oscillating CSF flow be-      scan also is shown to indicate the location of the blood flow
tween the cranium and spinal canal. During systole, arterial       velocities measurement. In the velocity encoded image, black
inflow is greater than venous and CSF outflows, resulting in a     pixels indicate flow in cranial direction (arterial inflow), and
small temporary increase in volume that in turn causes an          white pixels indicate flow in the caudal direction (venous
increase in pressure. Total arterial inflow to the cranium was     outflow). Graphs of total arterial (filled) and venous (open)
calculated by summation of the VFR through each of the four        VFR waveforms before and after decompression surgery for
vessels carrying blood to the brain (right and left internal       the same patient are shown in Figure 4. Arterial blood flow is
carotid arteries and the right and left vertebral arteries). Ve-   relatively unaffected by the surgery. An example of an MRI
nous outflow was obtained by summation of the flow through         scan used for quantitation of CSF flow is shown in Figure 5. A
the internal jugular veins and epidural, vertebral, and deep       sagittal T1-weighted MRI scan with the slice location for the
cervical veins when present. Pressure change during the car-       CSF flow measurements also is shown. Graphs of the CSF VFR
diac cycle was derived from the change in the CSF pressure         waveforms before and after surgery are shown in Figure 6. In
gradient. The CSF pressure gradient waveform was calculated        this patient, CSF VFR during the systolic phase is lower after
from the velocity-encoded MRI images of the CSF flow with          surgery compared with VFR measured before surgery. The

1346 | VOLUME 55 | NUMBER 6 | DECEMBER 2004                                                                    www.neurosurgery-online.com
                                                     CHANGES      IN INTRACRANIAL       COMPLIANCE       WITH    DECOMPRESSION SURGERY

                                                                           FIGURE 4. MRI-derived measurements of total arterial inflow (closed
                                                                           circles) and total venous outflow (open circles) for a patient before (A)
                                                                           and after (B) decompression surgery. The flow dynamic and the total arte-
                                                                           rial VFR before and after surgery were not affected by the operation.

FIGURE 2. Midsagittal T1-weighted MRI scans used to diagnose Chiari I
malformation and assess the level of herniation. A and B, a patient with
syrinx and without hydrocephalus before (A) and after (B) decompression
surgery. C and D, a patient with hydrocephalus before (C) and after (D)
decompression surgery.                                                     FIGURE 5. A, phase contrast MRI scan of systolic CSF flow through the
                                                                           spinal canal used for the calculation of the CSF VFR. B, anatomic midsag-
                                                                           ittal T1-weighted MRI scan showing the location of the low-velocity
                                                                           encoded cine phase contrast MRI scan used for CSF flow measurement.

FIGURE 3. A, phase contrast MRI scan of blood vessels to and from the
                                                                           FIGURE 6. The VFR waveforms of CSF before (A) and after (B) decom-
cranium. Black, arterial flow toward the cranium; white, venous flow
                                                                           pression surgery. A reduction in the amplitude of the CSF VFR waveform
from the cranium. B, blood vessel MRI scout image showing the location
                                                                           is observed after surgery. The lower heart rate during the postoperative
of the axial slice for blood flow measurements.
                                                                           MRI study (longer cardiac cycle period) may have contributed to this
lower VFR could have resulted from the longer cardiac cycle
(lower heart rate) during the postoperative study.
   Average intersubject values of maximum cord displace-                   0.05). The average CI increased 64%, from 6.9 before surgery to
ment, maximum CSF systolic velocity, maximum CSF systolic                  11.3 after surgery. The effect of surgery on ICC also was
VFR, CSF oscillatory volume, and CI, before and after surgery,             evaluated separately in the nine patients who did not have
are summarized in Table 1. On average, cord displacement                   hydrocephalus to eliminate its possible contribution to the
decreased by 21% after surgery, CSF systolic velocity de-                  overall ICC. The distribution of the intracranial CI for the nine
creased by only 2%, maximum CSF VFR decreased by 15%,                      patients with Chiari malformation without hydrocephalus is
and the oscillatory CSF volume decreased by 13%. None of the               shown in Figure 7. The preoperative average CI for these 9
changes in these parameters reached statistical significance. In           patients was slightly higher (7.7) compared with the average
contrast, the change in CI was statistically significant (P                for all 12 patients, possibly owing to the effect of the hydro-

NEUROSURGERY                                                                                       VOLUME 55 | NUMBER 6 | DECEMBER 2004 | 1347

   TABLE 1. Summary of magnetic resonance imaging-derived
   hydrodynamic parameters before and after
   decompression surgerya
                             Presurgery         Postsurgery
                                                                 P value
                            (mean SD)          (mean SD)

   Max. cord                 0.32     0.18      0.25     0.12      0.14
   displacement (mm)

   Max. mean CSF             1.60     0.45      1.56     0.64      0.39
   velocity (cm/s)
                                                                             FIGURE 8. Net transcranial A     V blood flow (closed circles) and the
   Max. CSF VFR             180.5     42.3     153.5     60.7      0.11      CSF flow waveforms to the craniospinal system (open circles), before
   (ml/min)                                                                  decompression (A) and after decompression (B). The CSF waveform fol-
   Oscillatory CSF           0.53     0.2       0.46     0.21      0.12      lows the A V waveform more closely before decompression.
   Volume (ml)
                                                                             ple of the net transcranial blood flow (arterial inflow minus
   Compliance index            6.9    4.6       11.3     5.8       0.01b     venous outflow [A      V]) as well as the CSF VFR waveforms
   (all patients)                                                            before and after surgery. Before surgery, the CSF waveform
   Compliance index     c
                               5.9    3.1       10.3     4.9       0.02b     “follows” the A V waveform more closely than it does after
                                                                             surgery. Because the net transcranial blood flow can be re-
     SD, standard deviation; max., maximum; CSF, cerebrospinal fluid; VFR,   garded as the driving force of the CSF flow, the less compliant
   volumetric flow rate.
     Statistical significance (P 0.05).
                                                                             intracranial compartment before surgery causes the CSF
     Excluding patient with Crouzon’s disease.                               waveform to follow the A V waveform more closely (5).

                                                                                This study expands earlier work performed by other
                                                                             groups, who used dynamic MRI measurements to quantify the
                                                                             effect of decompression surgery on local craniospinal junction
                                                                             hydrodynamics with measurements of cord displacement and
                                                                             CSF flow velocity. In the present study, changes in local hy-
                                                                             drodynamics also were measured and then compared with the
                                                                             change in ICC. Previous studies demonstrated that patients
                                                                             with Chiari malformation had higher tonsillar pulsations (22)
                                                                             and rates of cord displacement (18) compared with healthy
                                                                             subjects. Therefore, the finding of larger cord displacement
                                                                             before surgery in this study is in agreement with the results of
                                                                             earlier reports. It is possible that the difference in this param-
                                                                             eter did not reach statistical significance in the current study
FIGURE 7. Graph representing the change in CI after the decompression        because of the large intersubject variability in tonsillar herni-
surgery among patients with Chiari malformation without hydrocephalus.       ation (5–17 mm), which may have contributed to the variabil-
Possible reasons for the decrease in CI demonstrated in one patient ( )      ity in cord displacement.
and considerably larger values in another patient ( ) are described above       In the present study, mean CSF velocity, averaged for the
in the Discussion.
                                                                             entire cross sectional area, was used to represent CSF veloci-
                                                                             ties at the upper cervical spine. The average maximum CSF
                                                                             velocity was relatively unaffected by surgery. This finding
cephalous. The postoperative average CI in these 9 patients                  does not corroborate those in previous reports (7, 12), which
increased to 11.9. After surgery, an increase in ICC was dem-                had documented statistically significant increases in CSF ve-
onstrated in 10 of the 12 patients, was relatively unchanged in              locity at selected locations after decompression surgery.
one patient, and was significantly decreased in one patient.                 Bhadelia et al. (12) observed significant increase in velocity
The decreased ICC was demonstrated in a patient whose                        only in the anterior subarachnoid space below the foramen
symptoms persisted after surgery.                                            magnum, and Armonda et al. (7) reported increased velocity
   The increase in ICC after decompression is demonstrated by                only in the dorsal region at C2. These studies demonstrate the
the change in the relationship between the CSF and the net                   variability in velocity measurements and the strong depen-
transcranial blood flow waveforms. Figure 8 shows an exam-                   dence on the choice of the region of interest. Therefore, the

1348 | VOLUME 55 | NUMBER 6 | DECEMBER 2004                                                                             www.neurosurgery-online.com
                                                 CHANGES     IN INTRACRANIAL     COMPLIANCE          WITH    DECOMPRESSION SURGERY

relatively unchanged mean CSF velocity demonstrated in the              Although a decrease in the maximum amplitude of the CSF
current study can be attributed to the difference in the location    velocity at C3 was reported to correlate with the severity of the
of the measurement as well as in the inherent large variability      tonsillar herniation (24), MRI measurements of CSF velocity in
in the distribution of CSF velocities in the cervical spine region   the craniospinal junction are not widely used to assess the
(25).                                                                functional severity associated with Chiari malformation. The
   Unlike the local hydrodynamic parameters, the difference in       lack of correlation between the degree of the herniation and
ICC measured before and after surgery was statistically sig-         severity of symptoms further limits the usefulness of local
nificant, even with the relatively smaller number of patients        hydrodynamics parameters such as CSF velocity alone to char-
used in this study. There was a clear trend of increased ICC         acterize the pathophysiological characteristics of Chiari I
after surgery. Increased ICC was demonstrated in eight of the        malformation.
nine patients without hydrocephalus, and in two of the three            This study quantifies for the first time the effect decompres-
patients with hydrocephalus. The average CI measured before          sion surgery has on overall ICC. It further suggests that the
surgery in 11 patients (excluding the patient with Crouzon’s         change in ICC may have a greater significance than previously
                                                                     reported changes in local hydrodynamic parameters. This po-
disease) was 5.9. This value is considerably lower than the
                                                                     tentially establishes the CI derived from MRI measurements
average CI (8.24 2.42) previously reported in young healthy
                                                                     as a more sensitive indicator of the mechanical state of the
subjects (4). This finding supports the hypothesis by Hofmann
                                                                     system compared with CSF velocity alone. The finding of
et al. (18) that associates the increase in cord movement with
                                                                     decreased CSF VFR after surgery can be explained in relation
decreased ICC in patients with Chiari I malformation.
                                                                     to the increased ICC with decompression surgery. Because the
                                                                     craniospinal system is more compliant after surgery, it is able
                                                                     to accommodate the increased arterial inflow and blood vol-
Relationship between Outcome and Change in ICC                       ume during systole with a lesser amount of CSF displaced
   The increased ICC after decompression surgery in 10 of the        from the cranium to the spinal canal.
12 patients studied also was associated with improved clinical          This study supports suggestions by other investigators that
outcome. The abnormal decrease in the ICC after surgery              ICC has a potentially important role in the pathophysiological
occurred in a patient with approximately 10 mm of herniation         characteristics of Chiari malformation and the effect of decom-
and a syrinx extending from C2–C4. This patient presented            pression surgery (10, 18). In a report of failed decompression
with weakness in the right hand, which persisted after sur-          surgery treatment, Bejjani et al. (10) discuss the need to deter-
gery. An unusual thickening of the arachnoid membrane with           mine the compliance of the system before and after surgery to
adhesions was noted during surgery. It is possible that this         better understand the reasons for failure, which could be
abnormal thickening and the adhesions were further aggra-            addressed by additional treatment such as lumbar puncture or
vated by the surgery, and that this was the reason for the           ventricular shunting. Further study in a larger number of
decrease in ICC. Arachnoid scarring has been reported previ-         patients with negative outcome would be needed to substan-
ously by Bejjani et al. (10) as a potential cause for surgical       tiate a correlation between decreased ICC and negative out-
complications and failure. Thus, decreased ICC probably can          come. The ability to quantify ICC, noninvasively by MRI,
be associated with a poor surgical outcome in that patient. The      could potentially help identify the most effective treatment
                                                                     among the different decompression procedures currently of-
second patient in whom the change in ICC did not follow the
                                                                     fered. In addition, the MRI-based ICC measurement could be
trend also had hydrocephalus. An insignificant decrease in
                                                                     used to identify patients who may not benefit from decom-
ICC was measured after surgery, although most symptoms of
                                                                     pression surgery, thereby reducing failure rate and improving
headaches reported by this patient have resolved. The one
                                                                     overall outcome.
patient with abnormally high ICC values both before and after
surgery (Fig. 7) had syringomyelia and had undergone previ-
ous craniofacial surgery for the treatment of Crouzon’s dis-                                    REFERENCES
ease, which may explain the high CIs before and after surgery.
   The increase in CI after decompression surgery is visually        1. Alperin N, Lee SH: PUBS: Pulsatility based segmentation of lumens con-
                                                                        ducting nonsteady flow. Magn Reson Med 49:934–944, 2003.
evident from the relationship between CSF and the transcra-          2. Alperin N, Stelzig C: Does CSF flow reflect the mechanical state of the
nial net A V waveforms before and after surgery (Figure 8).             craniospinal system? Presented at the 7th International Society of Magnetic
The transcranial A      V blood flow can be regarded as the             Resonance in Medicine, Philadelphia, Pennsylvania, May 24–28, 1999.
                                                                     3. Alperin N, Lee SH, Loth F, Raksin PB, Lichtor T: MR-Intracranial pressure
input to the intracranial system and the CSF flow as the
                                                                        (ICP): A method to measure intracranial elastance and pressure
output. For a given pulsatile input inflow, the output flow             noninvasively by means of MR imaging—Baboon and human study. Radi-
would be less pulsatile (smoother) for the system with a larger         ology 217:877–885, 2000.
CI. The A V waveform shown is followed more closely by               4. Alperin N, Lee SH, Sivaramakrishnan A, Lichtor T, Blend M, Hemmati M:
                                                                        Determining the false positive rate of a new MRI-based method for simul-
the CSF waveform in the preoperative than the postoperative
                                                                        taneous measurement of total cerebral blood flow and ICP. Presented at the
example, indicating that the intracranial compartment is more           89th Scientific Assembly and Annual Meeting of Radiological Society of
compliant after surgery than before.                                    North America, Chicago, Illinois, November 30–December 5, 2003.

NEUROSURGERY                                                                                 VOLUME 55 | NUMBER 6 | DECEMBER 2004 | 1349

 5. Alperin N, Vikingstad EM, Gomez-Anson B, Levin DN: Hemodynami-                  Acknowledgments
    cally independent analysis of cerebrospinal fluid and brain motion ob-
    served with dynamic phase contrast MRI. Magn Reson Med 35:741–754,               We thank Ed and Gayle Labuda from the Charitable Fund of the Vanguard
    1996.                                                                           Charitable Endowment Program for financial support.
 6. Alzate JC, Kothbauer KF, Jallo GI, Epstein FJ: Treatment of Chiari type I
    malformation in patients with and without Syringomyelia: A consecutive
    series of 66 cases. Neurosurg Focus 11:Article 3, 2001.                                                  COMMENTS
 7. Armonda RA, Citrin CM, Foley KT, Ellenbogen RG: Quantitative cine-mode
    magnetic resonance imaging of Chiari I malformations: An analysis of
    cerebrospinal fluid dynamics. Neurosurgery 35:214–224, 1994.
 8. Arnett B: Arnold-Chiari malformation. Arch Neurol 60:898–900, 2003.
                                                                                    S  ivaramakrishnan et al. have provided a valuable addition
                                                                                       to the body of knowledge regarding the pathophysiology
                                                                                    and management of the Chiari malformation. They studied 12
 9. Badie B, Mendoza D, Batzdorf U: Posterior fossa volume and response to          patients, both before and after surgery, regarding magnetic
    suboccipital decompression in patients with Chiari I malformation. Neuro-
                                                                                    resonance imaging cine flow hydrodynamic characteristics at
    surgery 37:214–218, 1995.
10. Bejjani GK, Cockerham KP, Rothfus WE, Maroon JC, Maddock M: Treat-
                                                                                    the cervicomedullary junction. They studied spinal cord dis-
    ment of failed adult Chiari malformation: Decompression with CSF drain-         placement, volumetric flow rate, cerebrospinal fluid velocity,
    age observations in six patients. Acta Neurochir (Wien) 145:107–116, 2003.      and intracranial compliance. Most notably, they observed an
11. Bhadelia RA, Bogdan AR, Wolpert SM: Analysis of the cerebrospinal fluid         increase in intracranial compliance after decompression. This
    flow waveform using gated phase contrast MR velocity measurements.              suggests a global and far-reaching effect of symptomatic
    AJNR Am J Neuroradiol 16:389–400, 1995.
                                                                                    Chiari malformations, as well as providing insight into their
12. Bhadelia RA, Bogdan AR, Wolpert SM, Lev S, Appignani BA, Heilman CB:
    Cerebrospinal fluid flow waveforms: Analysis in patients with Chiari I
                                                                                    pathophysiology. Gradually, we are beginning to “under-
    malformation by means of gated phase-contrast MR imaging velocity mea-          stand” the Chiari malformation and its ramifications. Sivara-
    surements. Radiology 196:195–202, 1995.                                         makrishnan et al. have most certainly contributed to this body
13. Cama A, Tortori-Donati P, Piatelli GL, Fondelli MP, Andreussi L: Chiari         of knowledge with this article.
    complex in children: Neuroradiological diagnosis, neurosurgical treatment
    and proposal of a new classification (312 cases). Eur J Pediatr Surg 5[Suppl                                                    Edward C. Benzel
    1]:35–38, 1995.                                                                                                                 Cleveland, Ohio
14. Caviness VS: The Chiari malformations of the posterior fossa and their
    relation to hydrocephalus. Dev Med Child Neurol 18:103–116, 1976.
15. Elster AD, Chen MY: Chiari I malformations: Clinical and radiologic reap-
    praisal. Radiology 183:347–353, 1992.
                                                                                    T   he authors have studied a group of 12 patients before and
                                                                                        after decompression for the hindbrain herniation syn-
                                                                                    drome (Chiari I malformation) by use of magnetic resonance
16. Enzmann DR, Pelc NJ: Normal flow patterns of intracranial and spinal
                                                                                    imaging. Cine face contrast images were used to quantify cord
    cerebrospinal fluid defined with phase contrast cine MR-imaging. Radiol-
    ogy 178:467–474, 1991.
                                                                                    displacement, cerebrospinal fluid velocity, and volumetric
17. Guyotat J, Bret P, Jouanneau E, Ricci AC, Lapras C: Syringomyelia associ-       flow rate as well as the intracranial compliance. Their results
    ated with type I Chiari malformation: A 21 year retrospective study on 75       showed an increased overall intracranial compliance in 10 of
    cases treated by foramen magnum decompression with a special emphasis           12 patients. This information improves our understanding of
    on the value of tonsils resection. Acta Neurochir (Wien) 140:745–754, 1998.     some of the changes that occur with tonsillar impaction and
18. Hofmann E, Warmuth-Metz M, Bendszus M, Solymosi L: Phase contrast MR
                                                                                    intracranial compliance as well as cerebrospinal fluid flow
    imaging of the cervical CSF and spinal cord: Volumetric motion analysis in
    patients with Chiari I malformation. AJNR Am J Neuroradiol 21:151–158,
                                                                                    dynamics. However, the decision to operate on problematic
    2000.                                                                           patients and gain information with preoperative assessment
19. Levy WJ, Mason L, Hahn JF: Chiari malformation presenting in adults: A          requires further elucidation.
    surgical experience in 127 cases. Neurosurgery 12:377–390, 1983.
20. Menick BJ: Phase contrast magnetic resonance imaging of cerebrospinal                                                         Arnold H. Menezes
    fluid flow in the evaluation of patients with Chiari I malformation.                                                          Iowa City, Iowa
    Neurosurg Focus 1:Article 5, 2001.
21. Park JK, Gleason PL, Madsen JR, Goumnerova LC, Scott RM: Presentation
    and management of Chiari I malformation in children. Pediatr Neurosurg
    26:190–196, 1992.
                                                                                    T    his is an interesting submission. The authors have used
                                                                                         contemporary imaging technology to measure intracranial
                                                                                    compliance and assess the effects of suboccipital craniectomy
22. Pujol J, Roig C, Capdevila A, Pou A, Marti-Vilalta JL, Kulisevsky J, Escartin   and C1 laminectomy with duroplasty in the treatment of
    A, Zannoli G: Motion of the cerebellar tonsils in Chiari type I malformation
                                                                                    symptomatic patients with a Chiari I malformation.
    studied by cine phase-contrast MRI. Neurology 45:1746–1753, 1995.
23. Raksin P, Alperin N, Sivaramakrishnan A, Surapaneni S, Lichtor T:
                                                                                       Intracranial compliance was found to be substantially im-
    Noninvasive intracranial compliance and pressure from dynamic MR im-            proved after decompressive surgery in 10 of 12 patients. This
    aging of blood and CSF flows: Review of principles, implementation, and         work is a contribution to our knowledge base on this topic. We
    other noninvasive approaches. Neurosurg Focus 14:Article 4, 2003.               now can objectively assess patients before and after decom-
24. Shibuya R, Yonenobu K, Koizuma T, Kato Y, Mitta M, Yoshikawa H:                 pression to document whether our surgical procedure has had
    Pulsatile cerebrospinal fluid flow measurement using phase-contrast mag-        its intended effect (to improve intracranial compliance). I look
    netic resonance imaging in patients with cervical myelopathy. Spine 27:
                                                                                    forward to the study of changes in compliance and outcome to
    1087–1093, 2002.
25. Thomsen C, Stahlberg F, Stubgaard M, Nordell B: Fourier analysis of cere-       determine whether improved compliance truly translates into
    brospinal fluid flow velocities: MR imaging study—The Scandinavian Flow         better outcome. This type of assessment has broad potential
    Group. Radiology 177:659–665, 1990.                                             application for patients with increased intracranial pressure

1350 | VOLUME 55 | NUMBER 6 | DECEMBER 2004                                                                                    www.neurosurgery-online.com
                                                CHANGES       IN INTRACRANIAL        COMPLIANCE        WITH   DECOMPRESSION SURGERY

and reduced intracranial compliance from any cause, particu-            mations. This is a small group of patients, and it will be useful
larly when attempting to discern whether a previously suc-              to examine such studies in a larger group and in a repeated
cessfully treated patient is again symptomatic from scarring,           manner. A correlation with clinical outcome would be impor-
shunt dysfunction, tumor recurrence, or other cause.                    tant to our understanding of the physiology of Chiari I mal-
                                                                        formations and associated syringomyelia. Alteration in sys-
                                           Mark N. Hadley               tolic cerebrospinal fluid flow and velocity, cerebrospinal fluid
                                           Birmingham, Alabama          oscillatory volume, and intracranial compliance may be useful
                                                                        parameters to assess patient outcomes, and although local
T   he authors have examined in detail a small number of
    patients with Chiari I malformation by use of preoperative
and postoperative cine magnetic resonance imaging. Determi-
                                                                        hydrodynamic parameters were found to be less sensitive to
                                                                        decompressive surgeries, studies in additional patients may
nation of cerebrospinal fluid velocities can be complex, and            help clarify the subgroups in which such parameters are im-
consistency of such measurements can be difficult to achieve            portant. As a whole, this is an interesting paper that adds to
                                                                        our understanding and suggests new ways of using cine mag-
unless a clear location for comparison is identified.
                                                                        netic resonance imaging to assess the pathophysiology of
  Calculation of intracranial compliance is crucial to a variety
                                                                        Chiari malformations.
of problems in neurosurgery. The use of this technique by the
authors demonstrates a significant increase in intracranial                                                         Karin M. Muraszko
compliance after surgical decompression of Chiari I malfor-                                                         Ann Arbor, Michigan

                                 Country Dance (pastel and oil on canvas, 1921) by Picasso (courtesy of the
                                 Picasso Museum, Paris).

To top