Computed Tomography : A Brief Historical Perspective

Radiology, diagnostic imaging, organ imaging, call it what you will, logically may be regarded as part of the physical examination of a patient. It is a more extreme example of the process that started with the invention of the stethoscope, the process of augmentation of our senses. This process may be extended beyond our normal senses so we may now appreciate the electrical activity of the brain or the heart, using the electroencephalogram or electro-cardiogram. The very complexity of the technical processes involved, however, tends to make us see them as remote from the diagnostic relationship of patient and doctor. The diagnosis of diseases of the brain provides us with a unique example of the role of these "augmented senses". The inaccessibility and delicate nature of the brain restrict the diagnostic tools we may use. The presence of a rigid bony box, the skull, protecting the brain prevents us from using our senses directly to examine the organ. Indeed, if we could examine it in our usual manner, the brain tissue might be destroyed by the use of percussion or palpation. The relative opaqueness of bone to X-rays limits the use of almost all plain radiographs. It is this very limitation that has provided an incentive to devise techniques to demonstrate the brain without damaging the brain tissue. Copyright Royal Medical Society. All rights reserved. The copyright is retained by the author and the Royal Medical Society, except where explicitly otherwise stated. Scans have been produced by the Digital Imaging Unit at Edinburgh University Library. Res Medica is supported by the University of Edinburgh’s Journal Hosting Service: http://journals.ed.ac.uk ISSN: 2051-7580 (Online) ISSN: 0482-3206 (Print) Res Medica is published by the Royal Medical Society, 5/5 Bristo Square, Edinburgh, EH8 9AL Res Medica, 1981: 16-22 doi:10.2218/resmedica.v0i0.936 John Brown. Horae Subsecivae, Second Series. A. & C. Black, London Comrie, J.D. History of Scottish Medicine, Vol.2. The Wellcome Historical Museum, London. COMPUTED TOMOGRAPHY: A BRIEF HISTORICAL PERSPECTIVE Prof. JJK Best, MSc, MRCPf FRCR Department of Medical Radiology, University of Edinburgh Radiology, diagnostic imaging, organ imaging, call it what you will, logically may be regarded as part of the physical examination of a patient. It is a more extreme example of the process that started with the invention of the stethoscope, the process of augmentation of our senses. This process may be extended beyond our normal senses so we may now appreciate the electrical activity of the brain or the heart, using the electro-encephalogram or electro-cardiogram. The very complexity of the technical processes involved, however, tends to make us see them as remote from the diagnostic relationship of patient and doctor. The diagnosis of diseases of the brain provides us with a unique example of the role of these "augmented senses". The inaccessibility and delicate nature of the brain restrict the diagnostic tools we may use. The presence of a rigid bony box, the skull, protecting the brain prevents us from using our senses directly to examine the organ. Indeed, if we could examine it in our usual manner, the brain tissue might be destroyed by the use of percussion or palpation. The relative opaqueness of bone to X-rays limits the use of almost all plain radiographs. It is this very limitation that has provided an incentive to devise techniques to demonstrate the brain without damaging the brain tissue. The purpose of this article will be to document briefly the pursuit of effective "non-invasive imaging methods", to use the current jargon, that has led to computed tomography (Table 1). T A B L E 1: Methods of Visualising the Brain


University of Edinburgh
Radiology, diagnostic imaging, organ imaging, call it what you will, logically may be regarded as part of the physical examination of a patient.It is a more extreme example of the process that started with the invention of the stethoscope, the process of augmentation of our senses.This process may be extended beyond our normal senses so we may now appreciate the electrical activity of the brain or the heart, using the electro-encephalogram or electro-cardiogram.The very complexity of the technical processes involved, however, tends to make us see them as remote from the diagnostic relationship of patient and doctor.
The diagnosis of diseases of the brain provides us with a unique example of the role of these "augmented senses".The inaccessibility and delicate nature of the brain restrict the diagnostic tools we may use.The presence of a rigid bony box, the skull, protecting the brain prevents us from using our senses directly to examine the organ.Indeed, if we could examine it in our usual manner, the brain tissue might be destroyed by the use of percussion or palpation.The relative opaqueness of bone to X-rays limits the use of almost all plain radiographs.It is this very limitation that has provided an incentive to devise techniques to demonstrate the brain without damaging the brain tissue.
The purpose of this article will be to document briefly the pursuit of effective "non-invasive imaging methods", to use the current jargon, that has led to computed tomography (Table 1).

S kull Radiography
The first step in the pursuit was the in tro duction o f radiography to medical practice early in 1896.Wilhelm Roentgen announced the discovery o f X-rays in November 1895, and made public his first radiograph using a photographic emulsion plate a month later.It is doubtful if any technical innovation has been so rapidly introduced and commercially exploited.This could be achieved because the apparatus required to build an X-ray machine was available in many laboratories.
When Thomas Edison heard o f the discovery of X-rays, he turned his attention to the problem of visualising the rays.During January 1896 he screened several thousand fluorescing crystals hoping to find one better than the barium platinocyanide used accidentally by Roentgen in his discovery.Edison discovered that calcium tungstate fluoresced some ten times more brightly and he coated a sheet of card w ith the crystals as a fluorescent screen.This device fo r use in flu o ro scopy was advertised in Edison's mail catalogue later in the same year.Other technical improvements produced skull radiographs which showed tones w ith exquisite detail but could only detect intra-cranial structures if they were calcified (Fig. 1).Thus, radiographs could be used to detect (a) asymmetry o f volume of the cerebral hemispheres by displacement o f a calcified pineal gland, (b) calcified brain lesions, and (c) infer chronically increased intra-craniai pressure because o f erosion o f intra-cranial bony structures such as the dorsum sellae.Because o f the fa cility w ith which skull radiography visualised bone, it is still the firs t and best method to demonstrate fractures of the skull.

Pneumo-encephalography
Inspection o f radiographs o f the skull, chest or abdomen show that the structures th a t cast shadows o f easily recognised different radio densities are bones containing calcium, soft tissues containing chiefly water, structures containing fat and hollow organs containing air.The reason why air is so much less radio-dense than soft tissues is related to the fact that air is approximately 800 times less dense than water.It therefore might be expected that should air be introduced w ith in the cranial vault, it would be easily detected on a radio-graph.
The first demonstration o f the ventricular system by air came about, not as a result o f a planned clinical procedure, but from an accident.On 24th November, 1912, a 47-year old man was knocked down by a New Y ork tram, fracturing his skull and cutting his forehead above the right eyebrow overlying the fracture.Radiographs o f the skull confirmed the fracture and the man was in hospital fo r 12 days and then to o k his own discharge.Seven days later he was re-admitted complaining o f headache and vomiting.Examination showed papilloedema and a right extensor plantar response.Radiographs o f the head at the second admission showed the ventricular system clearly seen and "enormously dilated w ith what was probably gas or air.
A t surgery, a cranial decompression was performed.One lateral ventricle was tapped and "the removal o f the trochar was followed by tw o or three quick spits o f air and flu id , and then clear cerebrospinal fluid to the amount o f 8cc".The patient died one week later and at postmortem the dura over the frontal lobes was adherent to the fracture site and a connecting tract was traced from the right frontal sinus to the frontal horn o f the right lateral ventricle.It was postulated that air had been forced through the sinus and then into the ventricle when the patient sneezed or blew his nose.
What had happened was that the accident had introduced air, a contrast medium, into the brain which had been then detected radiographically.In retrospect it may seem a small step from this accidental demonstration of the ventricular system to the introduction o f a deliberate pro cedure fo r injection o f air to aid diagnosis.This does not seem to be the case, because it was not until 1918 that a surgeon, Walter Dandy, described the first injection o f air into the head and in his report he makes no mention o f the previous obser vations o f accidental intracranial air.Initially, air was put into the ventricles by direct puncture having made a burr hole in the skull vault.This procedure is called ventriculography.The in tro duction o f air into the subarachnoid space by lumbar puncture was subsequently described by Dandy in 1919 and is called pneumo-encephalo graphy.(Fig. 2).
Pneumo-encephalography was an important step in the evolution o f imaging of the brain because it was the first method that provided information about structures inside the cranial vault during life.It did not show the brain tissue directly, but allowed inference as to the location o f intracerebral masses by showing displacement o f the normal anatomy o f the ventricular systems and the cisterns.Unfortunately, the method causes considerable discomfort and is not w ith o u t risk.It is a method that has been largely superseded by later methods, but it should be remembered that it was the most commonly performed brain contrast procedure fo r more than four decades.

Cerebral Angiography
In January 1896 w ithin a month of the ann ouncement by Roentgen of the firs t radiograph of a hand, E. Hascheck, at the suggestion o f a colleague, O.T. Lindenthal, injected the brachial artery of a cadaver w ith Teichmann's mixture and demonstrated the blood vessels o f the hand.
The major landmark in the development of clinical angiography occurred in Lisbon, Portugal in 1926, where the Portuguese neurologist, Egaz Moniz, performed the firs t angiogram in a living patient.Moniz and his associate, Almeida Lima, appreciated that they must find a substance opaque enough to X-rays to be seen against the density of the calcium of the cranium and which was sufficiently non-toxic not to harm the patient.They tested several substances and strontium bromide was selected.Moniz and Lima proceeded to inject a 70% solution o f strontium bromide into the carotid artery o f a dog and produced a clear radiographic picture o f the cerebral arteries (Fig. 3).Although we no longer use the radio-isotope used by Moore, iodine 131, and we employ very much more sophisticated devices to detect the location o f the radioactivity w ith in the cranium, we still exploit the unique property o f intra cranial capillaries observed by Moore in his flu o r escein experiments.The capillaries of the cerebral vessels are impermeable to a wide range o f substances that pass readily through the capillaries in the rest o f the body.The 'Blood -Brain Barrier' as this property is often referred to , is frequently destroyed by adjacent disease and is absent in the vessels of tumours.Thus, normal brain does not 'take up' radio-isotopes and abnormal brain w ill therefore show as an area of radioactivity (Fig. 4).Radio-isotope brain scanning has been largely replaced by computed tomography except in vascular disease where it is used as a simple noninvasive form of arteriography (via an intravenous injection) to observe the pattern of blood flow , the dynamic radio-isotope brain scan.

Echo-encephalography
The concept o f measuring distance using sound is fam iliar to anybody who has observed a flash o f lightning and then counted the seconds to when they hear the peal o f thunder.The use of sound to measure distance is used by both bats and porpoises in a more complicated manner, they both generate the sound and listen to the echo and have evolved very accurate ranging and locating senses.
The sinking o f the steam-ship Titanic by an iceberg in 1912 precipitated the proposal to detect icebergs by sound echoes.More sophisticated underwater echo locating devices were developed during both World Wars and became the 'sonar' which is used today.
The use o f ultrasound -the name given to sound waves o f very high frequency and short wave-length well above the range audible to the human ear -to image internal organs in the human body is fam iliar to many people now because o f its use in estimating foetal maturation during pregnancy.One o f the earliest applications of ultrasound, however, was in a simple form to demonstrate the symmetry o f the volume o f cerebral hemispheres.A Swedish neurosurgeon, Lars Leksell, demonstrated in 1956 that a pulse of ultrasound introduced at right angles through the squamous temporal bone produced an echo from a midline soft tissue structure (Fig. 5).This echo is displayed on a television monitor and which may be photographed (Fig. 6).
F ig u re 6: A computed tomography scan, the transverse section demonstrating the frontal horns of both lateral ventricles and the cerebral tissue.
Many concepts of neurological disease have changed by virtue of computerised tomography scanning.The ability to detect disease now means that virtually all brain tumours may be detected and localised by computerised tomography during life.The impact on neuroradiological practice has been very great, investigations mentioned in this article which before the advent of computerised tomography were the tools which allowed the morphology of neurological disease to be demon strated have had their pattern of use significantly changed.Pneumo-encephalography is now in frequently performed, even in specialised neuro radiological centres.Radioisotope brain scans are now infrequently performed in centres with access to computerised tomography.Arteriography is still performed but the reason for its use has changed with emphasis turning from the demonstration of gross anatomy towards the demonstration of vascular abnormalities.
Computerised tomography scanning has be come such a crucial part of diagnostic neurology and neurosurgery that it has assumed the same role that conventional radiography has for the orthopaedic surgeon.It is impossible to conceive of a practice of modern neurology without access to computerised tomography.
Computerised Tom ography Distribution of the radio-density of tissues in the head.

F
ig u re 1: P lain la te ra l ra d io g ra p h o f th e s k u ll.T h is d e m o n s tra te s e x q u is ite b on e d e ta il o f th e cra n ia l v a u lt and base, fa c ia l bones and c e rvica l spine.T h e ce re b ra l tissue is n o t seen.

Figure 2 :
Figure 2: A fro n ta l view o f an a ir encephalogram showing air o u tlin ing both lateral ventricles.

Figure 4 :
Figure 4:A radio-isotope brain scan, lateral view show ing activity (the dark area) in the territory of the m iddle cerebral artery.The appearances suggest a cerebral infarct.
Figure 5: A n echoencephalogram .A n am plitude m odulated oscilloscope trace (repeated and inverted below) w hich show s 'sp ik e s' dem onstrating the squam ous tem poral echoes and a m idline 'sp ike'.