Original arlicle 14

Transcranial Doppler (TCD) Ultrasound as a Noninvasive Means of Monitoring Cerebrohaemodynamic Change in Hydrocephalus D. Goh J, R. A. Minns J, s. D. Pye2

Summary CerebraI blood flow velocity (CBFV) measurements by Transcranial Doppler (TCD) ultrasound were performed on 27 patients with hydrocephalus (Group I: neonates, Group 11: children). Simultaneous measurements of direct ICP and CBFV were performed during ventricular taps in 16 patients. There was a significant eorrelation between ICP and Resistance Index (RI = peak systolic-end diastoliclpeak systolic velocity) overall in Group 11 patients (p < 0.02) and in individual neonatal patients (p < 0.001). After ventricular taps and ventriculo-peritoneal shunting (17 patients) there was a consistent significant deerease in RI due to increased end diastolic velocity in all patients (p < 0.001). This suggests the RI is a reliable index of cerebrovascular resistance for serial monitoring in individual patients. There was an exponential pattern of decay in RI with CSF volume depletion (volume-f1ow velocity response) in 50/56 taps which allows calculation of a volumebuffering resenre before perfusion change occurred. Simultaneous ICP/CBFV monitoring during sleep may help to identify patients who are unable to compensate haemodynamicaIly during episodic increase in ICP and are a greater risk of ischaemic insult. TCD is a useful noninvasive technique of monitoring cerebrohaemodynamic change for initial assessment and further management of children with hydrocephalus.

Key words Transcranial Doppler ultrasound - Hydrocephalus - Resistance Index

Introduelion Progressive hydrocephalus causes secondary damage due to ischaemic effects from raised intracranial pressure or brain shifts. Spontaneous arrest can occur in some patients before significant ischaemic damage occurs and shunting procedures could then be avoided. Complications of shunting such as infection, blockage and overdrainage remain considerable problems. Repeated assessment of cerebral perfusion is thus important for optimal management from the initial evaluation at diagnosis, observation for progression or arrest through to monitoring for long-term shunt complications. Most currently Received July 13, 1991 Eur J Pediatr Surg I, Suppll (1991) 14-17 © Hippokrates Verlag Stuttgart . Masson Editeur Paris

widely available techniques of monitoring cerebral perfusion which are applicable to all age groups usually involve exposure to radioactive isotopes, ego Xenon 133, and thus repeated expo· sure for monitoring purposes is not justifiable particularly in young children. Transcranial Doppler (TCD) ultrasound techniques (1) provide a noninvasive means of measuring cerebral blood flow velocity (CBFV) in the basal cerebral arteries. We have assessed the use of TCD for monitoring cerebrohaemodynamic changes in children with hydrocephalus at initial evaluation as weIl as in those who have been shunted. This paper will report on changes in Doppler CBFV indices in relation to CSF volume manipulation and 10 spontaneous episodic ICP elevations associated with cerebral blood volume changes occurring in sleep. The Pourcelot Resistance Index (10), RI = (systolic-diastolic)/systolic velocity has been applied as an index of distal cerebrovascular resistance if RI increases due to a decrease in diastolic velocity. Archer et al (2) demonstrated a decreased RI with hypercapnia-induced increased cerebral blood flow in healthy neonates. Assuming a constant vessel diameter' change in mean flow velocity (MFV) may provide an index of change in mean volume flow (4).

Materials and methods Patients TCD examinations were performed on 27 patients; 11 were neonatal patients (Group I) and 16 patients were between 1 to 14.3 years of age (Group 11). Eleven of the Group 11 patients had existing ventriculo-peritoneal shunts at the time of assessment. The patients were divided into two groups because values of the normal range of RI in children differ from neonatal patients (5). (1) Doppler changes were assessed in relation to CSF volume manipulation: a) Ventricular taps were performed in 16 paüents (Group I: 10; Group II: 6). b) Ventriculo-peritoneal shunting/revision (VPS) procedures were performed in 17 patients (Groups I: 8; Group 11: 9). (2) Doppler changes were assessed in relation to spontaneous ICP increases due to cerebral blood volume chang('s in sleep: Eight sleep recordings were performed on 7 palients (Group 11).

Ventricular taps were performed when lhere \\'as a progressive increase in ventricular dilatation or when clinical signs and symptoms of raised ICP were present. Operative intervention was indicated when raised ICP levels confirmed blocked or inadequate shunt

Downloaded by: National University of Singapore. Copyrighted material.

IOept. of Paediatric Neurology, Royal Hospital for Siek Children, Edinburgh, U.K. and 2Dept. of Medical Physics, Western General Hospital, Edinburgh, U.K.

TCD Ultrasound as a NoninvasiL'e ll1eans oIi\lonitonng Cerebrohaemodynamie Change in Hydroeephalus

Methods ICP was directly measured through a 23-gauge butterfly needle inserted percutaneously in a frontal ventriculostomy reservoir using a nondisplacement strain gauge pressure transducer (Gaeltec Ltd) and simultaneously charted by a pen recorder. A separate reservoir is usually placed in all our patients to allow access for repeated ICP measurement and CSF drainage as required (8). If ICP levels were elevated for age, CSF was drained in 1 ml increments through a three-way tap arrangement till ICP levels were within normal range for age. CBFV was measured from the middle cerebral artery using pulsed-wave Doppler from a portable unit (Decoder, Doptek Ltd) with a 2 or 4 MHz probe. Doppler signals were recorded on a cassette deck while a remote control unit allowed voice-dubbing of ICP and CSF volume information onto tape. Doppler signals were subsequently played back and values of the RI, time-averaged MFV, peak-systolic (5) and end-diastolic (0) velocities were obtained by tracing the outline of the maximum velocity envelope using a light pen. Each calculated mean value represents the mean of - 10 stable waveforms of good quality.

1 a) CSFV measurements were made continuously throughout a ventricular tap \vith the corresponding ICP levels during the tap noted. Linear regression of ICP v€rsus HI was obtained and preand post-tap ICP and Doppler indices were compared using paired t -test.

lable 1

Intracranial pressure OCP) and Doppler indices pre- and postventrreular taps. I

Pre-tap

Post-tap

I

I

p value

I mean (sd)

I mean (sd) Group 1(Neonates: 10 patients) RI 0.78 (.05) 40.7(12) MFV (ern/sec) 74.9 (18) S (ern/sec) 15.8 (6) 0 (ern/sec) lep (rnrn Hg) 11.1 (3.5)

0.68 (.06) 48.2 (13) 79.0 (19) 24.6 (8) 4.9 (1.8)