Computed Tomographic Scanning of the Brain National Institutes of Health
Consensus Development Conference Statement
November 4-6, 1981
This statement is more than five years old and is provided solely for historical purposes. Due to the cumulative nature of medical research, new knowledge has inevitably accumulated in this subject area in the time since the statement was initially prepared. Thus some of the material is likely to be out of date, and at worst simply wrong. For reliable, current information on this and other health topics, we recommend consulting the National Institutes of Health's MedlinePlus http://www.nlm.nih.gov/medlineplus/.
This statement was originally published as: Computed Tomographic Scanning of the Brain. NIH Consens Statement 1981 Nov 4-6;4(2):1-7.For making bibliographic reference to the statement in the electronic form displayed here, it is recommended that the following format be used: Computed Tomographic Scanning of the Brain. NIH Consens Statement Online 1981 Nov 4-6 [cited year month day];4(2):1-7.
A Consensus Development Conference held at the National Institutes of Health on November 4, 5, and 6, 1981, reviewed scientific evidence related to computed tomographic (CT) scanning of the brain.
At NIH, the Consensus Development Conference brings together investigators in the biomedical sciences, clinical investigators, practicing physicians, consumer and special interest groups to make a scientific assessment of technologies, including drugs, devices, and procedures, and to seek agreement on their safety and effectiveness.
On the first two days of the meeting, a consensus development panel and members of the audience heard evidence presented on the following key issues:
- What are the indications and contraindications for employing CT scanning for diagnosis of intracranial lesions?
- How much radiation is delivered by presently available CT scan equipment?
- To what extent has CT scanning influenced the management of intracranial disorders?
- Has the availability of CT brain scanning influenced the use of other methods of imaging the brain?
- What is the practical limit of definition of CT scanning?
Members of the panel included physicians and scientists representing biomedical research, radiology, pediatric and adult neurology, neurosurgery, radiation therapy, radiation physics, critical care medicine, family practice, hospital administration, health economics, and other fields relevant to a discussion of CT.
CT is a remarkable new development in radiographic imaging which, in only 8 years, has transformed the diagnosis and much of the management of structural disease of the brain and its surrounding tissue. Presentation by experts from a variety of fields indicated that CT is a safe, accurate, and powerful tool in the primary diagnosis of, among other conditions, brain tumors, brain hemorrhages, effects of major head injury, and certain infections of the brain. Given its speed, accuracy, and low radiation dosage, CT has displaced a number of other radiologic diagnostic procedures, many of which are, in comparison, more uncomfortable, more dangerous, and more costly to the patient.
The panel considered the geographic distribution of CT scanners and the current patterns of their diagnostic usage. Two important considerations emerged. It appears that in the U.S. today, CT may not be sufficiently available for the public to derive the full benefit of this diagnostic tool. Evidence points to an insufficient number of instruments in some large metropolitan areas, in medically underserved areas, and in some sparsely populated regions where the prevalence of head trauma is high. In some instances, however, the indiscriminate use of CT has occurred in patients unlikely to have structural disease, resulting in displacement of patients for whom this technology is critical. Accordingly, this consensus report suggests appropriate criteria for the use of CT in present medical practice.
What Are the Indications for Employing CT Scanning as a Primary or Secondary Diagnostic Tool for Intracranial Lesions?
CT is the most useful diagnostic study available for a number of intracranial disorders. Its efficacy, particularly in the detection of traumatic and neoplastic lesions, is well established. There is evidence that CT has been a major factor in decreasing morbidity and mortality, especially in severe head injury and brain abscess. The presence of the former or suspicion of the latter present clear indications for CT.
CT detection of intracranial tumors is now so well established that the suspicion of any intracranial mass lesion from history or neurological examination constitutes a strong indication for this procedure. Spontaneous intracranial hemorrhage is also readily detected by CT and often represents another prime indication. Suspicion of arteriovenous malformation, hydrocephalus, herpes encephalitis, parasitic infestations, and progressive degenerative diseases of the brain generally constitutes an indication for CT.
For most cerebral vascular ischemic events, including transient ischemic attacks (TIA's), CT may not be required to make the diagnosis. The procedure, if available, is a helpful adjunct when using anticoagulant therapy. CT will clearly distinguish infarction from hemorrhage when that distinction must be made. CT will also detect lesions such as meningiomas or subdural hematomas. This discrimination is particularly important in the small percentage of patients with these conditions who have transient symptoms similar to TIA.
CT should not be employed as a routine screening procedure when a low diagnostic yield is anticipated. There is little indication for the procedure following minor head trauma, simple or periodic headache, syncope, or dizziness unattended by other neurological symptoms or signs. Most patients with headache should be considered for CT scanning only if the symptom is severe, constant, unusual, or associated with abnormal neurological signs.
In adults with seizure disorders the indications are variable. CT is essential when seeking a potential structural cause of complex partial (temporal lobe/psychomotor) or focal seizures and is indicated in any adult with a recent onset of seizures. The finding of an obvious clinical cause for seizures such as a withdrawal state or metabolic disturbance usually obviates the need for CT. If seizures remain uncontrolled and the initial CT was normal, the scan should be repeated at appropriate intervals.
In infants and children, CT is useful as a primary diagnostic tool in the evaluation of intracranial hemorrhage and mass lesions. CT may be used to evaluate patients postoperatively and those who have received radiotherapy.
In children, CT should be considered a primary diagnostic modality in evaluation of severe head trauma, other causes of increased intracranial pressure, undiagnosed coma, progressive focal neurological signs or symptoms, megalocephaly, and selected neurocutaneous syndromes. CT will also detect major congenital anomalies of the brain. It can detect perinatal intraventricular hemorrhage, but transfontanel ultrasonography is the preferred diagnostic procedure in infants under one year of age because of the portability of the ultrasonic equipment and the ease of performing this test.
CT is not necessary in evaluating the majority of children with developmental retardation, cerebral palsy, seizure disorders, or headaches, because the presence of a surgically treatable lesion is extremely low. The clinical situation must, in each case, be considered individually.
Are There Specific Contraindications?
CT is a remarkably safe diagnostic procedure and there are no absolute contraindications to its use. Its employment carries, nevertheless, potential hazards, as does any diagnostic procedure. These may include: separation of a critically ill patient from intensive medical and nursing care; an adverse response to contrast material; and a possible delay in applying emergency life-saving treatment to immediate life-threatening conditions such as acute arterial epidural hematoma. CT scanning facilities should have immediately available the staff and equipment to manage cardiorespiratory and neurological emergencies. The staff itself should be skilled in methods to prevent movement artifacts.
The use of contrast agents usually administered intraveneously, increases accuracy and may assist in characterizing lesions. The benefits of enhancement must be balanced against the increased risk to the patient should an adverse reaction occur. The total risk of the use of contrast materials is small and the incidence of severe reactions does not exceed 0.04%. Reactions, when they do occur, include allergic manifestations, hypotension, congestive heart failure, the development of renal insufficiency, and possibly adverse effects on the brain. The likelihood of complications is increased by advanced age, the presence of diabetes, cardiac, renal, or cerebral vascular disease and the administration of large doses of contrast material.
How Much Radiation Is Delivered During Use of Currently Available CT Scan Equipment, and How Is This Dosage Commonly Expressed?
CT irradiates a transverse section of the skull with a narrow x-ray beam, commonly 10 mm wide in equipment currently available. No single number can completely characterize the dose delivered to that section because the dose is not completely uniform throughout. For many purposes, the use of an average dose per section suffices. Typically, 10 adjacent transverse scan sections, each of 10 mm width are sufficient for examination of the entire brain. The average dose to the entire brain can be estimated from the average dose to each section. In addition to the average dose, the surface dose and doses of other points may also be stated.
The dose delivered to the brain depends upon individual machine characteristics and mode of operation which include the kilovoltage and milliamperage of the x-ray tube, the exposure time and tube current, the beam filtration, the x-ray target-to-patient distance, the CT detectors used, the scan angle, etc,. Because modern CT scanners utilize narrow x-ray beams with good radiation shielding, there is little dose delivered to tissues outside the imaged volume. Doses to sensitive organs, such as the thyroid gland and the gonads, which are well outside the useful beam, are much less than that received by the brain. The radiation dose to the lens of the eye is low if it is outside of the area scanned. Scanning should exclude the eye unless if cannot be avoided. Adequate maintenance and proper use of the scanner by trained technicians is necessary to insure that the radiation dose is the minimum required for the examination.
The unit by which dose is expressed is currently undergoing a transition from the commonly used rad (100 ergs/gram) to the gray (1 joule/kilogram). The relationship between the rad and the gray is: 1 gray = 100 rad. In this report, the unit centigray (cGy) will be used (one centigray = 1 rad). The average dose to the brain for a complete scan series ranges from about 1-10 cGy. This range is comparable to or less than the dose from many other procedures commonly used in diagnositc or dental radiology. Thus, as an x-ray procedure, routine CT does not deliver a particularly high radiation dose.
As is the case for all x-ray studies, particular consideration is needed in the examination of infants and children. The effects of repeated cumulative low-level radiation doses to the immature, developing brain (particularly from birth to two years of age) are as yet unknown. Special caution must be exercised in ordering multiple CT scans. In the infant, as noted earlier, ultrasound is often the preferred modality to monitor the status of abnormalities such as hydrocephalus.
Certain practices, including repeated scans, can increase the dose delivered. Special techniques--such as overlapping sections to detect small abnormalities, very thin sections to permit multiplanar reformatting*, slow scans to improve resolution, and dynamic scanning** --all increase the radiation dose. These studies may contribute useful information in specific cases but they are not recommended for routine use.
*A technique in which the data obtained from a series of scans made in one CT plane are reprocessed in the computer to produce images of sections in other planes through the head.
**Multiple sections taken at one location in rapid sequence to observe changing patterns of contrast enhancement.
To What Extent Has CT Scanning Influenced the Management of Intracranial Disorders, Such as Malignancy, Trauma, Vascular Anomalies, and Cerebrovascular Disease?
CT has had a major influence in the management of many intracranial disorders. It is a procedure entailing minimal discomfort and morbidity while producing a high degree of diagnostic accuracy. Lowered morbidity and mortality result from the decreased use of invasive diagnostic procedures.
Prompt employment of CT has improved the management of severe head trauma. The identification of traumatic intracranial mass lesions causing increased intracranial pressure has led to early surgical treatment of hematomas, the elimination of unnecessary surgical explorations, and more effective medical management. Evidence suggests that early surgical removal, i.e., within four hours, of acute subdural hematomas has decreased mortality and morbidity. In patients with head trauma sequential CT has permitted the identification of expanding intracranial subacute and chronic hematomas leading to an improved outcome.
The management and prognosis of patients with brain abscess have been substantially improved since the introduction of CT. The improvement is probably the result of earlier diagnosis permitting prompt institution of antibiotic therapy and more accurate determination of whether and when surgical intervention is needed.
In primary brain tumors, and in contrast to other diagnostic modalities, the use of CT has resulted in the detection of smaller lesions and more accurate localization; lower surgical morbidity and mortality and a decreased length of hospital stay have been the result. In metastatic brain tumors, CT identifies and localizes single and multiple lesions earlier than other techniques, thus permitting optimal treatment of both the metastatic and primary lesions. Postoperative complications due to hemorrhage or edema are identified more accurately, resulting in better treatment. Planning of radiation treatment of brain tumors is aided by CT as is follow-up evaluation after surgery, radiotherapy, and chemotherapy.
CT will usually differentiate between ischemic and hemorrhagic intracranial lesions. This distinction permits selection of patients for medical or surgical therapy. In subarachnoid hemorrhage, CT can obviate the need for lumbar puncture by the demonstration of subarachnoid blood in a high percentage of cases scanned soon after the hemorrhage. It can be helpful in localizing the source of the bleeding and may be useful in predicting development of vasospasm. CT can help to identify the ruptured aneurysm in patients harboring multiple aneurysms and sometimes shows unruptured asymptomatic or symptomatic aneurysms or arteriovenous malformations. In selected cases, this capability has led to better treatment of aneurysms and arteriovenous malformations and the prevention of late complications from these lesions.
Has the Availability of CT Brain Scanning Influenced the Use of Other Methods for Imaging the Brain?
The superior ability of the CT examination in the detection of intracranial and intracerebral disease processes has profoundly altered the use of several pre-existing radiographic methods for examining the brain. The modalities of skull roentgenography, geometric cranial tomography, cerebral angiography (arterial and venous), pneumoencephalography, positive contrast cisternography, radionuclide brain scanning, and ultrasonic echoencephalography have all decreased in use. Some of these examinations, notably the most invasive, those causing significant patient discomfort, or those with nonspecific diagnostic results have been affected more than others. Pneumoencephalography, radionuclide scans, and ultrasonic echoencephalography have almost disappeared from the neuroradiologic armamentarium where CT has been available. The utilization of skull roentgenography, geometric cranial tomography, and positive contrast cisternography has declined more recently as "modern" high resolution CT has become available. Such scanners provide bony anatomic information in addition to the unique low-density soft tissue data, available only with CT.
Although the use of cerebral angiography has not diminished to the same extent as the above-mentioned procedures, it is being used less frequently as a screening or diagnostic procedure.
The developments enumerated above have had a favorable impact upon health care delivery. CT scanning has often reduced risk and discomfort for patients. Data indicate that the costs of CT in diagnosis have been substantially offset by a reduction in costs resulting from discontinued procedures and the elimination of equipment needed to carry them out. In some cases net costs of diagnoses have even been reduced. The panel heard evidence that the use of CT in outpatient diagnosis has reduced hospital admissions and shortened hospital stay.
The success of CT scanning using x-ray has stimulated interest in tomographic imaging with other radiation sources. Similar computer techniques are being used to reconstruct images of the distribution of radioactive isotopes within the brain, procedures called emission computed tomography (ECT). Two forms of such techniques have been developed related to the type of radioactive materials and the detection system employed. These are single photon ECT (SPECT) and positron emission tomography (PET) scanning. SPECT and PET are able to measure functional variables such as local cerebral metabolism and blood flow that are not measured by CT. They disclose increased vessel permeability when present, which may also be detected on CT by contrast media enhancement. The spatial resolution of these techniques is less than that of CT, and as an imaging device the definition of the anatomic position of the measured function is limited with SPECT and PET. At present, these modalities are being applied in a limited number of institutions and their clinical role in the evaluation of stroke, epilepsy, and the metabolic aspects associated with mental disorders is the subject of ongoing research.
Recently, nuclear magnetic resonance (NMR), a technique using nonionizing forms of energy, has been used to produce sectional images of the human body. A strong magnetic field used in conjunction with a radiofrequency oscillating magnetic field, induces signals from atomic nuclei in tissues. These signals are reconstructed by a computer into sectional images in any plane in the head or body. The significance of the technique in clinical diagnosis remains to be determined. The potential advantages of NMR as compared to CT are the possibilities of providing improved imaging, and of identifying the chemical state of some important nuclei from which energy metabolism can be inferred at low resolution. In addition, the potential hazards associated with ionizing radiation are avoided by NMR. There are no known deleterious effects on tissues or organ systems from the magnetic fields and radiofrequency power densities presently used. The extent to which tissue characterization techniques will yield clinically useful information at practical financial costs has yet to be established.
What Is the Practical Limit of Definition and Resolution in CT Scanning That May Preclude Its Value in the Diagnosis of Brain Disease?
CT scanners are designed to detect morphologic abnormalities of the brain. In the opinion of the Panel, most contemporary scanners meet this goal with a high degree of accuracy. They discriminate among lesions depending on contrast and size. The lower limit of resolution is difficult to define, but some scanners are capable of detecting pituitary adenomas as small as 3 mm. Unfortunately, there is no way to know how many "negative" examinations are truly negative or simply due to limitations in the imaging method.
Current limitations in the quality of CT scans lie in spatial and contrast resolution and in scan speed. Dramatic improvements have occurred in these factors since 1973, but it is unlikely that this pace of improvement can be maintained. Improvement in contrast resolution, or the ability to detect a small signal against a noisy background, without increasing the radiation dose, is dependent on increasing dose efficiency which is approaching its theoretical limit. Spatial resolution is dependent on improvements in detector and semiconductor technology and can be expected to improve as these technologies advance. An increase in data obtained from selective sampling will lead to optimizing detection for a given signal to noise ratio; however, to obtain quantitative certainty commensurate with the improved sensitivity, radiation dose must be increased. Scan speed is largely limited by x-ray tube output and the ability of the x-ray tube anode to dissipate heat. New developments in x-ray tube design will lead to increases in scan speed.
Thus cost and the potential risk of increased radiation dose are the major practical constraints on further advancements in CT technology.
Consensus Development Panel
Fred Plum, M.D.
Chairman, Department of Neurology
Cornell University Medical College
The New York Hospital
New York, New York
Hillier L. Baker, Jr., M.D.
Professor of Radiology
William F. Collins, M.D.
Professor of Neurosurgery
Yale University School of Medicine
Yale-New Haven Hospital
New Haven, Connecticut
Edward R. Epp, Ph.D.
Professor of Radiation Therapy (Radiation Biophysics)
Harvard Medical School
Department of Radiation Medicine
Massachusetts General Hospital
Charles A. Fager, M.D.
Chairman, Department of Neurosurgery
Lahey Clinic Medical Center
Peggy C. Ferry, M.D.
Professor of Pediatrics and Neurology
Head, Section of Child Neurology
Department of Pediatrics
University of Arizona Health Sciences Center
C. Earl Hill, M.D.
Director, Family Practice Residency
University of Maryland School of Medicine
James E. Marks, M.D.
Associate Professor of Radiology
Mallinckrodt Institute of Radiology
Washington University School of Medicine
St. Louis, Missouri
Stuart O. Schweitzer, Ph.D.
Professor of Public Health and Director
Program in Health Planning and Policy Analysis
School of Public Health
University of California at Los Angeles
Los Angeles, California
Eugene L. Staples, M.H.A.
West Virginia University Hospital
Morgantown, West Virginia
Leslie M. Zatz, M.D.
Professor of Radiology
Stanford University Medical Center
Chief, Radiology Service
Veterans Administration Medical Center
Palo Alto, California
Dewey K. Ziegler, M.D.
Professor and Chairman
Department of Neurology
Kansas University Medical Center
Kansas City, Kansas
Jack E. Zimmerman, M.D.
Professor of Anesthesiology
Director, Intensive Care Unit
George Washington University Medical Center
National Institute of Neurological and Communicative Disorders and Stroke
National Center for Health Care Technology
Office of Medical Applications of Research