A Radioisotopic Test for Communicating

C. H. TATOR, M.D., PH.D., J. F. R. FLEMING,M.D., F.R.C.S.(C), R. H. SHEPPARD, M.D., F.R.C.P. (C), AND V. M. TURNER, B.A. Departments of Neurosurgery and Medicine, The Toronto Western Hospital, University of Toronto, Toronto, Canada

ECENTLY, there has been a rekindling of widespread use. In 1964, Di Chiro 6-8 reported interest in hydrocephalus, due largely a radioisotopic scanning technique for study- to increased awareness of this condi- ing CSF dynamics after the intrathecal tion in the adult. However, in adults, for two administration of radioiodinated human reasons, it is often very difficult to diagnose serum albumin (13qHSA). Our initial experi- symptomatic hydrocephalus. First, hydro- ence using this method has been reported. 15 cephalus in adults is usually secondary to or This paper describes its application to the associated with other intracranial lesions such diagnosis of communicating hydrocephalus. as tumors, meningitis, or subarachnoid hem- orrhage, and, in these cases, it is difficult to Method differentiate the symptoms and signs of the The tracer used was human serum albumin hydrocephalus from those due to the primary labelled with 113t (with a specific activity of lesion. Second, available tests, including mea- approximately 5 uc/mg).* It was diluted with surement of the (CSF) Elliott's "B" solution and injected into the pressure, dye studies, and radiography of the CSF by lumbar puncture with the patient in and subarachnoid spaces the lateral decubitus position, or by sub- using air or radioopaque material, often fail occipital puncture into the to establish whether the hydrocephalus is due with the patient sitting. For lumbar injection, to obstruction or atrophy. Measurement of the dose was 100 uc ~31IHSA in 10 ml, and for CSF pressure may not answer this question cisternal injection 100 uc in 5 ml. After injec- for, as Adams, et al., ~ have pointed out, the tion, the patient's activity and posture were CSF pressure may be normal in the presence not restricted. Scintillation scanning was per- of symptomatic hydrocephalus. formed using a Picker Magnascanner with a Radiographic studies may also be inconclu- 3-inch sodium iodide crystal. Anterior and sive. Hydrocephalus due to a complete block lateral scans were obtained 2 hours after within the ventricular system is readily diag- lumbar injection, and almost immediately nosed by air or positive contrast studies, but after cisternal injection. Repeat scans were when the block is incomplete and the ventricu- performed at varying intervals up to 48 hours lar system is only slightly larger than normal, after injection. the diagnosis may be difficult. With com- Because of the diminishing radioactivity municating hydrocephalus, the diagnosis is in the later scans due to physical decay of the reasonably certain if the air study shows a isotope, dispersion of the tracer in the sub- block in the subarachnoid space either in the arachnoid space, and absorption of tracer, basal cisternae or over the cerebral convexi- the sensitivity of the scanner was increased ties with failure of air to enter the sulci. for scans performed at later time intervals. However, if the cisternae and sulci are in- Therefore, the amount of radioactivity in completely blocked and a few of them do the scans is not strictly comparable, and this contain air, then diagnosis is difficult. Thus, factor must be considered when comparing in many cases, it may be impossible to es- the amount of radioactivity in the individual tablish whether symptomatic hydrocephalus scans of a series. The radioactivity of serial is present. Further methods for studying CSF blood samples was measured in a well-type dynamics are clearly needed. scintillation counter in order to study the rate Although several methods have been de- of transfer of tracer from CSF to blood. vised over the past 50 years, none has gained Received for publication August 14, 1967. * Obtained from Charles E. Frosst & Co., Montreal. 327 328 Tator, Fleming, Sheppard and Turner

Fx~. 1. Diagram of a lateral view of the normal intracranial subarachnoid space (left) and a normal lateral scan performed 45 minutes after the cisternal injection of 13qHSA (right). The letters are interpreted as follows: A = cisterna magna, B = cisterna pontis, C = cisterna interpeduncularis, D = suprasellar cistern, E = cisterna ambiens, F = quadrigeminal cistern, G = callosal cistern, H = cisterna lamina terminalis. The callosal cistern did not contain radioactivity in this scan.

Results through the foramen magnum into the spinal Normal CSF Scans. Diagrams of the normal subarachnoid space. From the cisterna pontis, intracranial subarachnoid space, and the there was diffusion laterally into the lateral corresponding scans in which the various cis- recesses of the cisterna pontis, but the main ternae can be identified, are shown in Figs. flow was directed anteriorly and superiorly 1 and 2. into the interpeduncular cistern. To reach When 131IHSA was injected into the sub- above the tentorium the tracer took two differ- arachnoid space, a consistent pattern of flow ent routes: posteriorly around the midbrain and distribution of tracer was observed. via the paired cisterna ambiens to reach subse- Following injection into the cisterna magna quently the quadrigeminal cistern, or ante- (Fig. 3), most of the tracer flowed around the riorly into the suprasellar cistern. Usually medulla to the cisterna medullaris and up into there was good visualization of all the basal the cisterna pontis. Some flowed downward cisternae on the first scans, which were com-

FIG. 2. Diagram of an anterior view of the normal intracranial subarachnoid space (left) and a normal anterior scan performed 6 hours after the lumbar injection of I*IlHSA (right). The large accumulation of radioactivity in the midline is due to superimposition in this view of many of the basal cisternae. The letters correspond to those in Fig. 1 ; I = Sylvian cistern, which is well demonstrated at this time.