Suggested standards for reports dealing with cerebrovascular disease☆
Article Outline
- Abstract
- Clinical classification
- Risk factors
- Diagnostic studies
- Results and complications of therapeutic intervention
- Appendix
- References
- Copyright
Abstract
The evaluation of clinical reports on vascular disease is often made difficult by variations in descriptive terms, clinical classification, and outcome criteria. In 1983 the Joint Council of the Society for Vascular Surgery and the North American Chapter of the International Society for Cardiovascular Surgery created the Ad Hoc Committee on Reporting Standards to address these problems and recommend solutions. Some general problems were addressed in the initial report dealing with lower extremity ischemia. This article concerns clinical standards for reports dealing with cerebrovascular disease, suggests a scheme for clinical classification, and recommends standardized reporting practices for grading risk factors, angiographic and other diagnostic findings, and the results and complications of therapeutic intervention. (J VASC SURG 1988;8:721-9.)
In the evaluation and treatment of cerebrovascular disease (CVD), vascular surgeons naturally focus on atheromatous disease at the carotid bifurcation and the role of carotid endarterectomy in its management. Many publications in this area use confusing and inconsistent terminology, making data comparisons difficult or impossible. Although this article focuses on carotid disease, an overall scheme whose framework encompasses all forms of cerebrovascular disease is proposed, to be more universally appealing and compatible with the needs of other specialists (e.g., neurologists).
Clinical classification
Cerebrovascular accident (CVA), stroke, transient ischemic attack (TIA), global ischemia, and many other terms that deal with CVD appear to hold different meanings for different authors. For example, to some the term “stroke” implies cerebral infarct, and the terms “hemorrhagic infarct” and “ischemic infarct” are often used to distinguish two different mechanisms of focal brain damage. However, an ischemic infarct may be caused either by a thrombotic or embolic event and even a hemorrhagic infarct is to some degree ischemic. Most authors know what they mean by such terms, but uniform definitions are lacking. Unambiguous terminology is obviously needed in this field. However, trying to attach precise definitions to all our current terms will still leave overlap and be incompatible because some terms describe the clinical event including the duration and severity of the associated neurologic defect, others identify origins of disease, and still others describe the pathologic state of the brain, the region affected, or both.
Studies of the natural history of CVD and its modification by therapeutic intervention require a framework of clinical classification that will permit separation of patients into meaningful subgroups. Until the early 1950s, most strokes were placed in the single general category of CVA. A growing understanding of the relationship between atherosclerotic disease of the extracranial and intracranial arteries and ischemic neurologic symptoms required development of more precise terminology. In 1960 Whisnant et al.1 proposed a classification for focal cerebral ischemia with the following categories: incipient stroke (intermittent or transient symptoms, arbitrarily defined as lasting less than 24 hours), advancing stroke (progressing or evolving symptoms), and completed stroke. The incipient stroke soon became known as TIA. This classification system became widely used and in time such additional categories as “crescendo” TIA and reversible ischemic neurologic deficit (RIND) were added. Other classification systems have been published by Natali and Thevenet2 in France and Vollmar3 in Germany.
The major limitation of all these clinically based classifications is that they do not provide for stratification of patients according to underlying disease. In an effort to overcome this problem, the National Insitute of Neurological and Communicative Disorders and Stroke4 (NINCDS) proposed a comprehensive classification covering six areas. These included (1) temporal profile (transient, prolonged, permanent deficit), (2) pathophysiologic mechanisms, (3) anatomy, (4) pathologic condition, (5) relevant clinical data (past history, physical findings, and test results including angiography and CT scan), and (6) patient status. Although the NINCDS classification is comprehensive, its very complexity makes it impractical for most clinical reports.
This problem stimulated Robert Courbier to organize a symposium in Marseilles in September 1984. Participants included 34 neurologists and vascular surgeons from Europe and the United States. Several problems with the prior classification systems were identified, including the following: (1) acronyms do not have the same meaning in different countries; (2) classifications proposed and used by vascular surgeons do not provide sufficient detail to satisfy neurologists and vice versa; (3) the proliferation of clinical categories, while providing greater precision, results in such small groups that statistically significant comparisons are often not possible; and (4) classification of patients with vague symptoms is difficult. Drawing on the deliberations at this symposium, Courbier5 proposed a two-level system consisting of a primary clinical classification with a secondary description of arterial and intracerebral lesions.
After the Marseilles symposium, Eugene Bernstein and Norman Browse further elaborated on this approach and developed a scheme analogous to the TNM staging system used for tumor classification. Their CHAT system recommends categories for current status and history, the responsible arterial lesion, and the pathologic status of the target organ, in this case, the brain or eye.6
We adopted a modification of the Bernstein-Browse proposal as the best approach to a broad-based clinical classification rather than a more restrictive one specific for the needs of vascular surgeons. Our choice of CHAT has been reinforced by the fact that a new NINCDS committee is currently reviewing the area of classification and is considering CHAT as the basis for their new system. Table I outlines the CHAT system and the categories chosen for clinical status, both current and prior historic events, and arterial and brain disease.
Table I. Proposed classification of stroke
| C Current status (<1 yr) | H History (>1 yr) | A Artery | T Target | |||
|---|---|---|---|---|---|---|
| Symptoms | Vascular territory | Symptoms vascular territory | Site | Pathology | Site | Pathology |
| 0—Asymptomatic | a—Carotid ocular (amaurosis fugax) | Same categories as current status | 0—No lesion | a—Arteriosclerosis | 0—No lesion | h—Hemorrhage |
| 1—Brief stroke TIA (<24 hr) | b—Carotid cortical | Current clinical | 1—Appropriate lesion | c—Cardiogenic embolus | 1—Appropriate lesion | i—Infarct |
| 2—Temporary stroke-with full recovery (24 hr to 3 wk) | c—Vertebrobasilar | 1-5 a-c | 2—Lesion only in another vascular pathway | d—Dissection | 2—Lesion only in another vascular territory | j—Lacunar |
| 3—Permanent stroke, minor (>3 wk) | d—Other focal | 3—Combined, appropriate lesion and lesion in another vascular pathway | e—Aneurysm | 3—Combined, appropriate lesion and lesion in another vascular territory | m—AVM | |
| 4—Permanent stroke, major (>3 wk) | e—Diffuse | Subscript s is used to indicate prior operation | f—Fibromuscular | n—Neoplasm | ||
| 5—Nonspecific dysfunction | r—Arteritis | q—Other | ||||
| 6—Improving stroke | t—Trauma | r—Retinal embolism | ||||
| 7—Fluctuating stroke | ||||||
| 8—Deteriorating stroke | ||||||
The clinical section of the classification is divided into current clinical presentation and past history (the C and H of CHAT). Although for most patients presenting symptoms will be of more recent onset, a 1-year cutoff was selected for the duration of current symptoms. A parallel category for past history separates the patients who were totally asymptomatic before their current presentation from those with prior cerebrovascular episodes in the same or in some other vascular territory. Although the current clinical history will usually apply to one hemisphere or one arterial distribution, the past history will likely be more global and include previous strokes in other territories. Additional subscripts could be used to designate the territory of the previous stroke (e.g., ipsilateral to current symptoms... contralateral...vertebrobasilar), but this appears to complicate unnecessarily the basic CHAT system and is not recommended for routine use at this time. The major categories for clinical presentation include asymptomatic, brief stroke or TIA (less than 24 hours), temporary stroke (full recovery in 24 hours to 1 month), and permanent stroke (signs or symptoms lasting longer than 1 month), separated into major and minor deficits. In addition, there are categories for changing stroke and for nonspecific dysfunction. The past history section has the same categories as clinical presentation except that there is no need for the changing stroke categories. Each of the symptom categories is modified by a subcategory to define the vascular territory involved (i.e., carotidocular, carotid-hemispheric, vertebro-basilar, other focal or diffuse). In addition, a subscript s will be used to identify a previous cerebrovascular operation.
The changing stroke categories (6 through 8) were specifically included to identify those patients in whom the outcome of the current clinical episode is unknown before treatment (e.g., the patient who has carotid endarterectomy during a stroke-in-evolution). The category of nonspecific dysfunction is regrettably necessary for those patients who do not fit the other, more precise, classifications. Not only would their placement in other categories be difficult to decide on with consistency, but their inclusion there might obfuscate the evaluation of treatment for these more discrete symptoms.
Grading the severity of residual neurologic deficits after strokes, either before or after carotid endarterectomy, has a practical value in assessing the risk and the benefit of that operation. For many purposes strokes can be classified as major or minor, simply based on whether or not patient independence is maintained. If a more refined definition of neurologic complications is required, we recommend the Neurologic Event Severity Scale (Table II), adopted from the randomized trial of extracranial-intracranial bypass.7
Table II. Neurologic Event Severity Scale*
| Severity grade | Impairment† | Neurologic symptoms | Neurologic signs |
|---|---|---|---|
| 1 | None | Present | Absent |
| 2 | None | Absent | Present |
| 3 | None | Present | Present |
| 4 | Minor, in one or more domains | Present | Present |
| 5 | Major, in only one domain | NA‡ | NA |
| 6 | Major, in any two domains | NA | NA |
| 7 | Major, in any three domains | NA | NA |
| 8 | Major, in any four domains | NA | NA |
| 9 | Major, in all five domains | NA | NA |
| 10 | Reduced level of consciousness | NA | NA |
| 11 | Death | NA | NA |
| *Adapted from Stroke 1985;16:397-406. Reprinted by permission. †Impairment in the domains of swallowing, self-care, ambulation, communication, and comprehension. If independence is maintained despite the impairment, it is classified as minor; if independence is lost, it is classified as major. ‡Neurologic signs and symptoms are integrated into the higher grades of impairment. | |||
The classification of arterial involvement includes the location and nature of lesions identified. Although more complex schemes designed to provide an accurate anatomic localization and gradation of severity of the arterial lesions were considered, a simpler approach is endorsed of identifying whether a lesion in the arterial distribution of the affected region of the brain, or in some other location, was present or absent. The arterial disease is thus classified as follows: (0) no lesion; (1) lesion location appropriate to symptoms; (2) lesion only in another vascular pathway; and (3) 1 and 2 combined. Category 1 is only used when the vascular abnormality identified is considered to be the source of the symptoms. The arterial disease is coded as arteriosclerosis, cardiogenic embolus, dissection, aneurysm, fibromuscular dysplasia, arteritis, trauma, or other. In most cases the clinician can identify the most likely pathologic condition; however, in exceptional situations it may be necessary to code more than one category of disease. Most patients treated by vascular surgeons fall in the atherosclerosis or fibromuscular dysplasia groups, but this system includes the broader range of patients seen by neurologists and neurosurgeons. However, in reports dealing exclusively with atheromatous disease of extracranial arteries, it would be appropriate to categorize the arterial lesions by the same criteria developed for the diagnostic studies used to identify them. These are described in a subsequent section of this article and include grades of stenosis from 0 to 4 and degrees of ulceration from 0 to 3.
The section dealing with brain disease uses a scheme analogous to that for the arterial lesion, indicating whether or not a lesion could be identified by angiography, CT scanning, magnetic resonance imaging, position emission tomography, radionuclide scan. or other (future) diagnostic methods, in the appropriate territory of the brain to explain the presenting symptoms. It also allows for the possibility of a lesion in another territory or coexisting lesions in both. Specific pathologic entities include hemorrhage, infarct, lacunae, arteriovenous malformation, neoplasm, retinal embolus, and other.
This classification system not only accommodates the wider spectrum of disease encountered by other specialists but has the advantage over onedimensional schemes of being able to deal with patients in whom the combination of clinical status and brain lesion creates categorization problems. A good example of this would be a patient without symptoms who has a high-grade carotid stenosis and two infarcts in the ipsilateral hemisphere shown by CT scan. In most schemes, the disease of this patient might be classified simply as an “asymptomatic” carotid stenosis and be grouped with other heretofore inactive and less dangerous carotid lesions. Another example is a patient with a brief left hemispheric stroke and an ulcerated left carotid artery who has a history of right hemispheric infarct occurring after total occlusion of the right carotid artery. Other classification systems would either not recognize the contralateral lesions or overcategorize the patient because of them.
The CHAT classification scheme, when used in its entirety, may be more comprehensive than needed for many vascular surgeons, particularly with regard to the categories of arterial and brain disease that are included. However, it is not necessary to use the entire CHAT system routinely; one is not required to separate patients by artery or target. Some reports may only use the current presentation portion of the classification. However, adhering to the guideline will help to maintain uniformity in reporting and provide more meaningful analysis. When data are not available, the appropriate part of the scheme is not used. The Appendix includes examples of classification.
Risk factors
General risk factors for development of atherosclerosis are also related to the severity and rate of progression of cerebrovascular disease. Although many alternative grading systems have been proposed, none is generally accepted. For the sake of uniformity, we have adopted the scheme proposed by the Subcommittee on Reporting Standards for Lower Extremity Ischemia8 (Table III).
Table III. Risk factor categories
| 0 | 1 | 2 | 3 | |
|---|---|---|---|---|
| Drabetes | None | Adult onset, no insulin | Adult onset, insulin controlled | Juvenile onset |
| Tobacco use | None (or abstinence >10 yr) | None currently, abstinence 1-10 yr | Current, >one pack/day (or abstinence >1yr) | Current ≥1 pack/day |
| Hypertension | Diastolic <90 mm Hg | Easily controlled, single drug | Requires two drugs | > Two drugs or uncontrolled |
| Hyperlipidemia | Cholesterol and triglycerides within normal limits for age | Mild elevation, diet controlled | Types II, III, or IV requiring strict diet control | Require drug control |
| Cardiac | Asymptomatic, normal ECG | Asymptomatic, remote MI (> mo), occult MI by ECG | Stable angina, controlled ectopy, asymptomatic arrythmia, drug-compensated congestive failure | Unstable angina, symptomatic or poorly controlled ectopy or dysrythmia, poorly compensated failure, MI <6 mo |
| Renal | No renal disease, creatinine <1.5 mg/dl, clearance >50 ml/min | Creatinine 1.5-3.0 mg/dl Clearance 30-49 ml/min | Creatinine, 3.1-5.9 mg/dl, clearance 15-30 ml/min | Creatinine >6.0 mg/dl, clearance <15 ml/min, dialysis or transplant |
| Pulmonary | Asymptomatic, normal chest x-ray film PFT >80% of predicted | Asymptomatic or mild dyspnea on exertion, mild x-ray parenchymal changes, PFTs 65%-80% of predicted | Between 1 and 3 | Vital capacity <1.85 L, FEV1 <1.2 L or <35% of predicted, maximal voluntary ventilation <28 L/min or <50% of predicted, PCO2 >45 mm Hg, supplemental O2 required, pulmonary hypertension |
It is not necessary for every report to investigate and characterize each of these factors, but the categories listed provide a scale of severity from normal to severe disease for each factor, should such detailed characterization be appropriate. In some reports the presence or absence of such risk factors need not be mentioned or their incidence may be simply noted to provide an overall perspective of the patient population. However, claims that treatment variables affect outcome (e.g., variations in operative technique or the use of antithrombotic drugs) deserve detailed characterization of those risk factors also known to affect outcome in the treatment groups being compared.
Diagnostic studies
Angiography
Although problems arise when severity of arterial disease is defined by contrast arteriography, this technique remains the current “gold standard” by which other diagnostic methods are measured. The severity of occlusive lesions should be classified according to the percentage of diameter reduction measured on the view that shows the greatest extent of luminal reduction. Although the physiologic effect of a stenosis is directly related to reduction in cross-sectional area, this parameter cannot be measured directly and an estimate from a two-dimensional view may introduce substantial error. Some investigators prefer to measure the residual lumen at the point of greatest stenosis as a more critical end point, but this method introduces scaling errors in attempting to correct for magnification. Digital subtraction studies represent an even greater problem because there is no satisfactory method to determine the scale of the images. The percentage in diameter reduction should be measured relative to the outflow vessel. Thus the maximal diameter is measured from a normal segment of the internal carotid artery beyond the upper extent of disease (Fig. 1).
Fig. 1.
Method for measuring percentage of stenosis.
Categories of the diameter of stenosis should be clinically relevant. Physiologic studies have shown that substantial hemodynamic disturbances (flow and pressure decreases) begin to occur at approximately 60% stenosis. Several recent clinical studies have reported that the natural history of carotid stenoses worsens significantly when stenosis exceeds 80%. Therefore we recommend that less than 20% narrowing be included under “normal,” with stenoses of 20% to 59% being “mild,” 60% to 79% “moderate,” 80% to 99% “severe,” and 100% “occlusion.” For computer purposes, these could be ranked from 0 to 4. A second aspect is the gradation of surface irregularity and ulceration. A simple qualitative classification is as follows: (0) normal smooth vessel surface, (1) small irregularity (less than 2 mm in depth), (2) moderate (between 2 and 4 mm), (3) severe irregularity (more than 4 mm deep, multiple sites or lesions containing an intraluminal filling defect).
Noninvasive tests
Many clinical reports on the natural history of carotid bifurcation disease or steresults of noninvasive tests. Too often these reports involve terminology and methodology that are not well defined or standardized. Most reports do not contain information concerning the test criteria employed or evidence of validation of the method in the vascular laboratory used. Many authors simply quote results published by others with the presumption that similar studies in their laboratory are equally reliable. Thus conclusions are reached and recommendations are made on the basis of unsupported assumption that the noninvasive techniques used are highly accurate.
Another subcommittee of the SVS/ISCVS Committee on Reporting Standards that is dealing with noninvasive testing should be presenting its specific recommendations in the near future. Until then, as a minimum, all articles in which conclusions are based primarily on the results of noninvasive tests should include (1) the specific criteria used for interpretation of results and (2) validation data for the technique in the respective laboratory. Test accuracy should be reported in terms of sensitivity, specificity, and positive and negative predictive values.
Results and complications of therapeutic intervention
Benefit of therapeutic intervention is measured by the improvement on the natural history of the underlying disease, in this instance primarily in stroke prevention. The risk of such intervention is traditionally measured in operative mortality and permanent neurologic morbidity. Patency after operation of extracranial arteries is important but it does not serve as the primary yardstick of success as it does for lower extremity arterial reconstruction. Furthermore, there are problems in applying the basic categories of “improved,” “no change,” or “worse” to patients without symptoms who, even after successful relief of a critical stenosis, can hardly be considered improved. Nor can reduction in the number or frequency of TIAs be accepted as improved. Finally, improvement after carotid endarterectomy is difficult to guage for those with brief, temporary, changing, or even permanent stroke, in whom spontaneous improvements is relatively frequent.
Therefore the success or failure of such intervention is best gauged by the absence or occurrence of stroke in the distribution of the artery operated on, the relief or recurrence of preoperative symptoms, or both. It is not appropriate to apply an overall postoperative stroke rate because it may include strokes in the contralateral, untreated carotid artery. Although antithrombotic drugs might be expected to give overall protection, operation can only be expected to prevent stroke or relieve symptoms in the distribution of the operated artery. The reason for the additional qualifier, “recurrence of preoperative symptoms,” is exemplified by operating on a significant stenosis contralateral to hemispheric hypoperfusion symptoms that are primarily related to an ipsilateral carotid occlusion. Failure of such operation would include either a stroke in the distribution of the vessel operated on or recurrent symptoms in the opposite hemisphere. Thus when surgical intervention is directed toward one of a combination of occlusive lesions responsible for hypoperfusion symptoms, recurrence of those symptoms, as well as ipsilateral stroke, constitutes treatment failure. This policy will also allow inclusion of diagnostic errors in which the arterial lesions operated on was, in retrospect, not responsible for the symptoms. Another common example of attempted relief of ischemic symptoms the localization of which is outside the normal territory of distribution of the artery operated on, is carotid endarterectomy performed on a significantly narrowed carotid artery but for relief of vertebrobasilar insufficiency caused by either occlusive lesions in the posterior circulation or a steal phenomenon occurring after subclavian artery occlusion. This should be dealt with according to the same principles mentioned above; the operation should be credited for the relief of appropriate preoperative symptoms in the other territory but it also should be blamed for postoperative neurologic events in both territories.
The converse of the stroke or symptom-free designation of patients operated on are those dying or suffering postoperative neurologic sequelae. Traditionally for carotid endarterectomy, this risk has been expressed by the quoting of mortality and neurologic morbidity rates. The practical value of these two yardsticks is attested to by their widespread adoption. Without meaning to detract from this, the committee believes that further detail should be supplied in articles accepted for publication in a scientific journal. These two figures may not reflect the true postoperative stroke rate if (1) some patients died of stroke or (2) permanent neurologic deficit was produced by injury to cranial nerves. The former underestimates the rate of postoperative neurologic deficit, the latter overestimates it. Where known, asymptomatic occlusive events should also be reported. Table IV presents a list of reportable complications.
Table IV. Reportable complications of operations to treat extracranial cerebrovascular disease
| Mortality |
 Nonstroke-related |
| stroke-related |
| Permanent neurologic morbidity |
| Stroke, major |
| Stroke, minor |
| Cranial nerve injury |
| Transient neurologic morbidity |
| Brief stroke |
| Temporary stroke |
| Cranial nerve injury |
| Asymptomatic occlusive events |
| Occlusion |
| Stenosis; technical |
| Recurrent stenosis |
Postoperative central neurologic complications should be reported with the clinical classification scheme (brief, temporary, or permanent). Permanent strokes should be stratified into major or minor or graded according to the Neurologic Event Severity Scale. Use of the severity scale is helpful where claims are made of improved neurologic outcome for one over other treatment modalities, especially in patients with preoperative neurologic deficits. Postoperative strokes may be separated further into the following categories: operative stroke, where the deficit is detected at the first opportunity to examine the patient after operation; perioperative stroke, with worsening detected after the initial evaluation and up to 30 days; and late postoperative stroke, occurring after 30 days. As usual, death from any cause during this same 30-day period must be counted as an operative death; deaths from stroke must be separated from those from other causes.
Cranial nerve injuries should be identified separately and listed by nerve and whether the deficit is temporary or permanent. In studies that focus on cranial nerve complications the type of surveillance used should be described. For example, many cases of vocal cord dysfunction may be missed unless routine laryngoscopy is employed.
Angiographic complications represent one of the risks to the patient having surgical treatment and may contribute to the overall morbidity rate. Although the data are not always available, it is advisable to describe the incidence of major complications of cerebral angiography including transient and permanent neurologic deficits as well as significant cardiac abnormalities, peripheral thromboembolic problems, and death.
Ideally, follow-up information on the patient's status should be obtained through formal examination by a physician, although information on occurrence of major stroke or death can be determined by telephone call or questionnaire. Life-table analysis of survival and of symptom-free intervals is the best method for presenting follow-up data. The report of the Subcommittee on Reporting Standards for Lower Extremity Ischemia contains an excellent discussion of the application of this method.8 Recommended guidelines for life-table analysis include the following: (1) data should be provided in tabular form in addition to the summary graph; (2) the total number of patients in each time interval should be noted on the graph; (3) points on the graph should be joined by steps rather than a smooth line; (4) any data beyond the point where the standard error exceeds 0.10 should be omitted or represented in a different way (e.g., as a dashed line) so as to emphasize the poor reliability of estimates beyond this point. Comparison of different life-table curves should be by the log rank test or a similar method that analyzes the data over the entire period of comparison.
The incidence of recurrent stenosis remains an important topic in the study of carotid endarterectomy. Data on symptomatic recurrent stenosis or occlusion are usually readily obtained and reported, but identification of asymptomatic recurrent stenosis is more difficult to assess and usually depends on noninvasive studies. Any formal prospective study of carotid recurrent stenosis should include some reliable means of assuring technical adequacy (e.g., completion angiography or operative ultrasound study), and baseline noninvasive studies should be obtained as early as feasible after operation to distinguish true recurrence of a residual stenosis.
Appendix
At first look the CHAT classification appears complex with a large number of permutations; however, in practice most surgical reports will deal with a limited number of combinations. In their initial use of the system Hye et al.6 found that 79% of 402 patients having carotid endarterectomy were without symptoms or had history of brief strokes, categories C0 or C1. Taking all patients who had been without symptoms within the last year the simplest classification would be C0. Because by definition there is no vascular territory to specify, there would be no letter modifier. Further subdivision of the group without symptoms is ach eved with the remaining elements of the system. A patient who never had any previous symptoms on the contralateral side would be C0-H0, whereas history of contralateral TIA 2 years before would be C0-H1. The nature of the earlier symptoms can be specified; for example, with prior amaurosis fugax the classification would be C0-H1a. Prior carotid operation would be indicated with the modifier s: C0-H1as. Distinguishing between ipsilateral and bilateral carotid disease on angiography or ultrasound is achieved by using C0-A1 as against C0-A3. The patient without symptoms with a silent infarct detected by cerebral scanning is coded C0-T4 instead of C0-T0 (or just C0 if no CT scan data were available). Strokes of differing severity (including TIAs) are classified C1 through C8. Different levels of risk are described with the modifiers to determine the presence or absence of prior episodes and types of arterial disease and target organ (brain) findings. For example, the patient with a left hemispheric stroke, prior left retinal infarct, and silent disease of the right carotid artery is coded as C3-H3-A3. Where desired, the vascular territory of the stroke can be added (e.g., C3b-H3a-A3). It should be noted that the current presentation category refers to symptoms occurring within the past year, so that the patient who had a major stroke 2 years ago and has no further new symptoms would be classified C0-H4. A more complete description of this same person might be C0-H4b-A3a-T4.
Categories C6 to C8 are reserved for those patients in whom the natural outcome of the stroke process is not known. In patients who do not have surgery or some other specific intervention, the end result of the stroke eventually is known and they can be reclassified, usually as C2, C3, or C4. However, if a patient with a fluctuating stroke pattern has an endarterectomy within 24 hours of symptom onset, it is not known whether he would have been classified as a TIA (C1) or as a major permanent stroke (C4); therefore, he is assigned to the C7 group permanently. Because the prognosis for patients with changing stroke who are improving or deteriorating differs, the additional categories C6 and C8 are included.
References
- . Appraisal of the current trer d toward surgical treatment of occlusive cerebral vascular disease. Med Clin North Am. 1960;44:857–885
- . Chirurgie des artères carotides et vertebrales dans leur segment extra-cranien. Paris: Masson; 1973;
- . Reconstructive surgery of the arteries. Stuttgart: G Thieme Verlag; 1980;
- . A classification and outline of cerebrovascular disease. Stroke. 1975;6:564–616
- In: Courbic R editors. Basis for a classification of cerebral arterial diseases. Amsterdam: Excerpta Medica; 1985;
- . Evaluation of a new classification of cerebrovascular disease (CHAT). Am J Surg. 1987;154:104–110
- . The international cooperative study of extracranial/intracranial arterial anastomosis (EC/IC Bypass Study): methodology and entry characteristics. Stroke. 1985;16:397–406
- Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg. 1986;4:80–94
☆ Reprint requests: J. Dennis Baker, MD, Dept. of Surgery, UCLA School of Medicine, Los Angeles, CA 90024.
PII: 0741-5214(88)90081-X
doi:10.1067/mva.1988.avs0080721
© 1988 Society for Vascular Surgery and International Society for Cardiovascular Surgery, North American Chapter. Published by Elsevier Inc. All rights reserved.
