Parametric imaging of cerebral vascular reserve

EuropeanJournalof

Nuclear Medicine

Original article

Parametric imaging of cerebral vascular reserve 2. Reproducibility, response to CO 2 and correlation with middle cerebral artery velocities A. Ross N a y l o r 1, M a l c o l m V. M e r r i c k 2, J a m e s M. S l a t t e r y 4, A l p Notghi 2, C o l i n M. F e r r i n g t o n 2 and J. D o u g l a s M i l l e r 3 1 Department of Surgery, Royal Infirmary of Edinburgh, 2 Department of Nuclear Medicine and 3 Department of Clinical Neurosciences, Western General Hospital, and 4 Department of Medical Statistics, University of Edinburgh, Edinburgh, UK Received 11 July and in revised form 15 September 1990

Abstract. We report the reproducibility and response to change in end-tidal CO2 of a new method of quantifying regional mean cerebral transit time (MCTT) compared with the reproducibility and CO2 reactivity of middle cerebral artery (MCA) blood flow velocities measured using transcranial Doppler ultrasound. Within the range of end-tidal CO2 which could be achieved in conscious subjects breathing spontaneously, hemispheric MCTT, peak MCA velocity and mean MCA velocity showed a linear relationship with end-tidal CO2. After correction to a standardised end-tidal CO2, the coefficients of variation were 5.7% for hemispheric MCTT, 6.3% for peak MCA velocity and 6.8% for mean MCA velocity. Under the conditions of this study, MCA blood flow velocity was proportional to the reciprocal of MCTT, which in turn represents the ratio of blood flow to blood volume. Although the two methods appear to provide similar information, measurement of MCTT is quicker to perform, is less observer-dependent, provides regional information, uses conventional equipment present in most nuclear medicine departments and is less subject to problems associated with patient movement. K e y words: Cerebral transit time - Transcranial Doppler - Cerebral vascular reserve

Eur J Nucl Med (1991) 18:259-264

Introduction

[

provided by these more complex and expensive techniques, it is clear that if an evaluation of cerebral haemodynamics is to become a routine investigation in patients: with cerebrovascular disease, more readily accessible methods must be developed. One step in this direction is to adopt some of the principles identified by tomographic methods and to use less complex techniques to apply them more widely, an example being the concept of cerebral vascular reserve (CVR). Recent work using positron emission tomography (PET) has suggested that the best indicator of the extent to which cerebral blood flow is being maintained: by vasodilatation is the ratio of cerebral blood flow t o cerebral blood volume. This is another expression for the reciprocal of mean cerebral transit time (MCTT) (Gibbs et al. 1984). We have recently described the theory and methodology of a new way of quantifying M C T T (Merrick et al. 1990). If such a method were to prove reproducible and to respond in a predicted manner t o changes in end-tidal CO2, then it could identify impaired CVR prior to carotid surgery, in the evaluation of patients presenting with acute stroke and in the identification and monitoring of patients with vasospasm after subarachnoid haemorrhage. This facility would then be! available to any hospital with a nuclear medicine department. The aims of the current study were to evaluate the reproducibility and CO2 reactivity of our new methodl of quantifying MCTT and to compare the results with I the reproducibility and CO2 response of middle cerebral I artery (MCA) velocities using transcranial Doppler sonography.

It is sometimes suggested that non-tomographic methods of studying cerebral blood flow should be regarded as obsolete. While valuable information, particularly with regard to the pathophysiology of stroke, has been

Materials and methods

Offprint requests to: M.V. Merrick

Twelve healthy, conscious volunteers (age range 23-54, median 33 years) underwent five comparative studies of MCTT and MCA

© Springer-Verlag 199i

260 velocity of blood flow. Three reproducibility (basal) studies were performed at 15-min intervals, followed by two further studies during hyperventilation and rebreathing. A further subject underwent an identical series of studies, but only MCTT was measured. Endtidal CO2 was continuously monitored using an infra-red analyser with analogue and digital display through a multichannel patient monitor (Kontron Supermon). Permission for this study was obtained from the Lothian Area Surgical Ethics Committee and the UK Administration of Radioactive Substances Advisory Committee (ARSAC), and informed consent was obtained from each volunteer. Transit times. Full details of the methodology, theory, in-vitro validation and practical aspects of the new method have been described elsewhere (Merrick et al. 1991). Regional MCTT was measured following a peripheral intravenous injection of non-diffusible tracer with the subject in a supine, head-extended position over a 10-inch field-of-view gamma-camera equipped with a high-sensitivity collimator. Because the half-life of Tc 99m sodium pertechnetate precludes repeat examinations within a 12-18 h period, 500-1000 MBq (15-30 mCi) of gold 195m (tl/2=30 s) was employed as the nondiffusible tracer for this study. In addition to dead-time correction, a decay correction was applied using the standard formula:

ventilation was achieved, recordings of MCA velocity, MCTT and the respective end-tidal CO2 were made. In one subject, the MCA signal was lost during maximal hyperventilation. All volunteers were allowed to recover fully, and once the end-tidal CO2 was back to normal, repeat studies were made during maximal rebreathing through a closed circuit rebreathing system. In each case, the study was terminated when the subject indicated by a pre-arranged hand signal that the maximal tolerable limit had been reached. Statistical analysis. The ranges of MCTT and of MCA velocity obtained during the basal studies were not normally distributed (Filliben's Test; Filliben 1975). As a consequence, non-parametric tests were employed. MCTT, velocity and inter-hemispheric differences have therefore been expressed as the median with 95% confidence limits. Comparative analysis within individuals was performed using the Wilcoxon Signed Rank Test (for paired data), assuming the null hypothesis. The coefficient of variation was determined in the standard manner. All statistical analyses were performed using the MINITAB package on an IBM compatible personal computer.

Results

Ct--~ C o . C - k t

where Ct is the count rate at time t, Co the count rate at the start of measurement, k the decay constant and e the exponential constant. Sequential 32 × 32 matrix frames were acquired at a rate of 3 per second for lmin, by which time the examination was complete. The only mean transit time (MTT) which can be measured is that from the start of injection to exit from the field of view of the detector. Thus, the MTT measured with a gammacamera viewing the head is that from arm to internal jugular vein (IJV), not the MTT through the cerebral circulation. The latter can, however, be calculated as the difference between the MTT from arm to IJV and that from arm to aortic arch (Merrick et al. 1991). A second detector was therefore positioned over the sternum in order to obtain the MTT from arm to aortic arch. Middle cerebral artery velocities. Measurement of blood flow velocity in the MCA was used as the comparative technique to quantification of MCTT as both methods provide an index of velocity flow in the cerebral circulation. Peak and mean velocity of blood flow in the main trunk of the MCA were measured using the Eden Medical Elecronics (EME TC2-64B) transcranial Doppler in the manner described by Aaslid et al. (1982). The intracranial portion of the internal carotid artery and its division into the anterior cerebral and middle cerebral arteries was insonated using a 2-MHz hand-held probe, usually through the posterior temporal window. The depth setting was then progressively reduced until the MCA signal alone was identified. Recordings were made when the velocity signal was maximum, at which the angle of insonation between the ultrasound beam and the MCA is assumed to be between 0° and 30°, thus keeping the maximum error below 15% (Aaslid et al. 1982). Basal reproducibility studies. Prior to commencing any study, the volunteer lay quietly until a stable end-tidal CO2 was obtained. The main trunk of the MCA was insonated and MCA velocities and MCTT recorded. This sequence was repeated at 15-min intervals until three basal studies had been obtained. Hyperventilation and rebreathing. The MCA was insonated and the volunteer instructed to hyperventilate. When maximal hyper-

The median hemispheric transit time (hMCTT) during the b a s a l studies was 5.5 s (5.3-5.7 s) at a n e n d - t i d a l C O z of 5.0% ( 4 . 9 % - 5 . 2 % ) . F i g u r e l a shows the n o r m a l distrib u t i o n of r e g i o n a l M C T T at a n e n d - t i d a l CO2 of 4.9%. T h e n o r m a l p a t t e r n consists of s h o r t e r M C T T s at the front o f the brain, with l o n g e r M C T T s p o s t e r i o r l y . T h e r e is close i n t e r - h e m i s p h e r i c M C T T s y m m e t r y ; the m e d i a n m o d u l u s o f i n t e r - h e m i s p h e r i c M C T T difference d u r i n g b a s a l studies was 0.2 s (0-0.6 s). D u r i n g h y p e r v e n t i l a t i o n (Fig. lb), the m e d i a n h M C T T i n c r e a s e d to 11.1 s (10.5-12.0 s) at a n e n d - t i d a l CO2 of 2.2% ( 2 . 1 % - 2 . 4 % ) . T h e m e d i a n m o d u l u s of i n t e r - h e m i s p h e r i c M C T T difference d u r i n g h y p e r v e n t i l a t i o n was 0.6 s (0.4-0.9 s). D u r i n g r e b r e a t h i n g (Fig. lc), the m e d i a n h M C T T fell to 4.0 s (3.6-4.4 s) at a n e n d - t i d a l C O a o f 6.8% ( 6 . 4 % - 7 . 1 % ) . T h e m e d i a n m o d u l u s of i n t e r - h e m i s p h e r i c M C T T difference d u r i n g r e b r e a t h i n g was 0.2 s (0.1-0.3 s). T h e r e was a l i n e a r r e l a t i o n s h i p b e t w e e n h M C T T a n d e n d - t i d a l (ET) CO2 (Fig. 2), with a g r a d i e n t of - 1.47 s/ 1% c h a n g e in E T C O E (95% confidence i n t e r v a l = - 1 . 3 1 to - 1 . 6 3 , r = 0 . 8 4 , P < 0 . 0 0 0 1 ) . W h e n the r e s p o n s e to c h a n g e in e n d - t i d a l CO2 was p l o t t e d as left h M C T T a g a i n s t right h M C T T , there was a n e x t r e m e l y high c o r r e l a t i o n ( r = 0 . 9 4 , P < 0 . 0 0 0 1 ) . A l l subjects r e s p o n d e d in a s i m i l a r m a n n e r , a n d w h e n h M C T T was p l o t t e d a g a i n s t e n d - t i d a l C O a o n a n i n d i v i d u a l basis, there was n o significant difference in i n d i v i d u a l g r a d i e n t s of slope. T h e indiv i d u a l c h a n g e in h M C T T b e t w e e n b a s a l a n d h y p e r v e n t i l a t o r y studies was statistically significant (P < 0.0001), as was the i n d i v i d u a l c h a n g e in h M C T T b e t w e e n b a s a l a n d r e b r e a t h i n g studies (P < 0.0001). U s i n g the m e a n g r a d i e n t of slope d e r i v e d f r o m Fig. 2, all r e p r o d u c i b i l i t y studies were c o r r e c t e d to a s t a n d a r d i s e d e n d - t i d a l COE o f 5.0%. T h e coefficient of v a r i a t i o n was 5.9% for left h M C T T a n d 5.5% for right h M C T T .

261

201 15

._

~g~o "o ~

=

':" o-.. :...

,

1 . ". . , ,

0

:',.~ iI I

:h

I

I

1

I

I

I

2

3

4

5

6

7

End tidal CO 2

(%)

I

Fig. 2. Relationship between hemispheric MCTT and end-tidal (ET) COz. The regression equation is: MCTT = 14.1- 1.47.ETCO2 (r = 0.87, P < 0.0001) Table 1. Normal values for peak and mean velocity during basal and during hyperventilation and rebreathing studies

Peak velocity (cm/s) Mean velocity (cm/s) End-tidal COz (%)

Basal

Hyperventilation

Rebreathing

94 (90-96)

61 (54-68)

134 (123-145)

60 (58 62)

30 (25-36)

93 (8(~99)

5.0 (4.8-5.2)

2.3 (2.1-2.7)

6.6 (6.3-7.0)

Values expressed as the median with 95% confidence limits in parentheses

Fig. la-c. Changes in regional mean cerebral transit time (MCTT) during basal (a), hyperventilation (b) and rebreathing (e) studies. Each colour shade represents an increment of 1 s in time from 0 to 20 s. Green colour in b denotes MCTTs in excess of 20 s The aortic transit time (ATT), measured in order to correct for dispersion of the bolus (Merrick et al. 1991), also provides an index of cardiac output (Thompson et al. 1964). During the basal studies, the median A T T was 6.1s (5.6-7.4s). During hyperventilation, the median A T T fell to 3.9 s (3.3-4.5 s), while during maximal rebreathing, it fell to 5.4 s (4.7-6.4 s). The difference between the median A T T during the basal and hyperventilatory studies was significant (P = 0.002), as was that between basal and rebreathing studies (P=0.005).

The median values for peak (PV) and mean (MV) M C A blood flow velocity during each phase of the study are detailed in Table 1. There was a linear relationship between peak M C A velocity and end-tidal CO2, with a gradient of + 15.5 cm/s per 1% change in ETCO2 (95% confidence interval= 12.8-18.1, r=0.82, P