Surgical Risk as Related to Time of Intervention in the Repair of Intracranial Aneurysms

Full access

It is generally agreed that the surgical risk in intracranial aneurysm is closely related to the patient's condition at the time of surgery, as well as to other factors such as age, associated disease, and the location of the aneurysm. Many criteria have been proposed1,2,5 for the determination of surgical risk; some are based upon whether or not the patient is “conscious,” others upon the number of days that have passed since the last hemorrhage, and still others upon the patient's age. We have felt that the intensity of the meningeal inflammatory reaction, the

It is generally agreed that the surgical risk in intracranial aneurysm is closely related to the patient's condition at the time of surgery, as well as to other factors such as age, associated disease, and the location of the aneurysm. Many criteria have been proposed1,2,5 for the determination of surgical risk; some are based upon whether or not the patient is “conscious,” others upon the number of days that have passed since the last hemorrhage, and still others upon the patient's age. We have felt that the intensity of the meningeal inflammatory reaction, the severity of neurological deficit, and the presence or absence of significant associated disease should provide the best clinical criteria for the estimate of surgical risk.

Our modification of Botterell's classification3 has been applied to 275 consecutive cases of intracranial aneurysm treated by the faculty and resident staff of the Ohio State University and affiliated hospitals over a 12-year period (Table 1).

TABLE 1

Classification of patients with intracranial aneurysms according to surgical risk

Category*Criteria
Grade IAsymptomatic, or minimal headache and slight nuchal rigidity.
Grade IIModerate to severe headache, nuchal rigidity, no neurological deficit other than cranial nerve palsy.
Grade IIIDrowsiness, confusion, or mild focal deficit.
Grade IVStupor, moderate to severe hemiparesis, possibly early decerebrate rigidity and vegetative disturbances.
Grade VDeep coma, decerebrate rigidity, moribund appearance.

Serious systemic disease such as hypertension, diabetes, severe arteriosclerosis, chronic pulmonary disease, and severe vasospasm seen on arteriography, result in placement of the patient in the next less favorable category.

In this series, almost all cases were graded at admission and again just prior to operation. It is recognized that such classifications are arbitrary and that the margins between categories may be ill defined. We are, nevertheless, of the opinion that a fairly sharp differentiation is possible among patients who have few or no meningeal signs, patients who have well-defined meningeal signs but no neurological deficit, and patients who show neurological malfunction. Associated disease was sufficient to change the graded risk when it was unequivocally present and, in the judgment of the surgeon, severe enough to influence the patient's operative or postoperative course. We did not consider it necessary for the patient to have recovered completely from all symptoms and signs of the hemorrhage to be considered an optimal or Grade I risk.

Table 2 shows the relative distribution of cases upon admission, classified according to this method. The largest single group of patients were those classified as Grade II, and the second largest group were those classified as Grade III.

TABLE 2

Classification of aneurysm at time of admission

GradeNumberDeathsPer Cent
I61711
II882326
III792937
IV352571
V1212100
Total2759635

Assuming that neurological deficit indicates arterial spasm, ischemia, and brain edema and that under such conditions the intracranial contents are more vulnerable to manipulation, and then further assuming that patients without neurological deficit would be best able to tolerate surgery, we established the following policy with regard to operation.

First, patients graded I and II were taken to surgery as soon as a diagnosis could be made, preferably within 24 hours of admission. Second, with two exceptions, patients graded III or below were treated conservatively until they improved to Grade I or II.

The first exception was that patients who had multiple, repeated, bleeding episodes were operated upon at Grade III or lower; the second was that patients who appeared to have an intracranial hematoma threatening life were operated upon at once and the hematoma evacuated, with or without definitive repair of the aneurysm.

Obviously, with our criteria, serious systemic disease or marked vasospasm will delay operation. However, neither the patient's age nor the site of the aneurysm are used in determining the grade of risk. Furthermore, the number of days elapsed since the last hemorrhage is not considered relevant. Figure 1 shows that the majority of good risk patients were operated upon within 3 weeks of their last hemorrhage, many in less than 1 week.

Fig. 1.
Fig. 1.

Time between hemorrhage and surgery for Grade I, II, and III patients. The vertical bars represent the number of cases operated on each post-bleeding day (measured from the last known hemorrhage) for those cases that had bled. The grading was done at the time of operation. Note the number of cases operated within the first week, and within the first three weeks in each grade. Delay in surgery in Grades I and II beyond the first day or two was usually incurred prior to admission to our institutions.

This paper represents a retrospective analysis of the results of the policies described above.

Results of Early Intervention in Good Risk Patients

In patients admitted to the hospital at Grade I or II (Table 3) the mortality was 20%, whereas in patients who reached the operating room for any procedure whatever at Grade I or II, it was 14%. The difference in mortality seems to be due to a number of instances of early fatal rebleeding. Fourteen of the 15 nonsurgical deaths occurring in patients who were admitted at Grade I or II were caused by rebleeding (Table 4). The other death was due to a dissecting aneurysm following angiography.

TABLE 3

Results of early intervention in good risk patients

GradeGrade at Time of AdmissionGrade at Time of Surgery
No. of PatientsNo. of Deaths%NumberNo. of Deaths% 
I617119055.5 
II882326691725 
I & II14930201592214 

TABLE 4

Deaths in nonoperated good risk patients (Grade I and II)

Cause of DeathNo. of Cases
Rebleeding14
Angiographic1
Total15

Analysis of these 14 deaths caused by rebleeding shows that 10 of them died after being in the hospital over 24 hours. In six of these instances we might have proceeded more rapidly to complete diagnostic procedures and surgery, in three instances diagnosis was incomplete because of a failure to obtain vertebral angiography or because of misinterpretation of carotid angiography, and in only one, who bled after 24 hours, was the aneurysm considered inoperable. In the other four instances, rebleeding occurred so soon after admission that we did not feel there was time to make a diagnosis and perform surgery.

We conclude, therefore, that the major cause of death in patients admitted in good condition is rebleeding, and that delay in diagnosis and treatment should be kept to an absolute minimum.

Another source of concern regarding the much higher mortality for patients graded II, as compared to those graded I, is shown by Table 5. The mortality rate is 22% among patients graded II at intracranial surgery and only 1.4% for those graded I, an average of 11%. The major difference between Grades I and II, as noted previously, is the presence of significant meningeal reaction or serious associated disease. Analysis of the Grade II intracranial operative deaths indicates that eight appear to be due to cerebral infarction caused by vasospasm. This seems to relate the danger of infarction to the inflammatory reaction about the vessels. This observation suggests that further delay prior to surgery may be advisable in some patients admitted at Grade II, if the meningeal reaction is particularly intense.

Delayed Intervention in Grade III Patients

The second major policy, that of delayed surgical intervention in patients showing neurological deficit, was subjected to similar analysis. With few exceptions, patients operated upon at Grade III were either early in the series, victims of repeated hemorrhage, or apparently suffering from life-threatening hematoma. The operative mortality was high (Table 5).

TABLE 5

Grade of aneurysm at time of surgery related to mortality (intracranial ligation only)

GradeNo. of PatientsNo. of Deaths%
I7111.4 
II581322 
III20840 
IV7343 
V000 
1562516 

Theoretically, the chief danger in delay of surgery is that rebleeding will become the major cause of death, as it is in the Grade I and II groups. However, this did not prove to be the case. Upon analyzing the patients who were Grade III at admission, we found that a significant number of them improved to better grades and were operated upon with a satisfactory survival rate (Table 6). Of the 23 patients admitted at Grade III who were not operated upon, seven survived in a badly crippled state and 16 died. Of these 16 nonoperative deaths, late rebleeding occurred in only four instances; the principal cause of death was cerebral infarction (Table 7). Earlier operation, which might have prevented the four cases of rebleeding, could be expected to have a survival rate of about 50%, but could not have possibly produced the salvage that delayed intervention has provided, at least in our hands (Tables 5 and 6). The waiting policy permitted 55% of the patients admitted at Grade III to reach a level at which surgical hazards were much less than they would have been on admission (Table 8).

TABLE 6

Analysis of patients rated Grade II at time of admission

Operative CategorySurvivorsDeaths
Operated at Grade I121
Operated at Grade II228
Operated at Grade III55
Operated at Grade IV30
Operated at Grade V00
Not operated upon716
Total4930
TABLE 7

Cause of nonoperative deaths among patients rated Grade II at time of admission

Cause of DeathNo.
Infarction8
Rebleeding after more than 24 hours4
Incomplete diagnosis2
Rebleeding in less than 24 hours1
Hemorrhage into metastasis1
Total16
TABLE 8

Significance of waiting policy regarding patients rated Grade III at time of admission

Status of Patient%
Improved to Grade I20
Improved to Grade II35
Remained at Grade III11
Deteriorated to lower grades34

Discussion

This paper presents a method for the evaluation of risk and selection of the optimal time for surgical intervention. Figure 2 summarizes all our cases. Of the 78 nonoperated cases that constitute the difference between the first and second columns, 58 died and 20 survived. The 20 survivors were not operated on for a number of reasons, including refusal of surgery (5), inoperability of the lesion (3), and overwhelming residual neurological deficit (12 cases).

Fig. 2.
Fig. 2.

Classification of patients by grade at admission, at surgery for all surgical procedures, and at surgery for intracranial ligation alone. The mortality for each classification is shown as a black wedge.

The 41 cases represented by the difference in the second and third columns were subjected either to cervical carotid ligation, intracranial trapping procedures, or external reinforcement procedures. The third column shows only those cases in which the aneurysm was obliterated by intracranial operation without sacrifice of parent vessels.

Our bias toward eventual surgical treatment is clear, but our over-all results are based upon a variable period of conservative therapy directed toward minimizing the risks of ischemia or rebleeding. The results indicate that a positive approach to the surgical repair of these lesions is warranted.

We feel that certain conclusions drawn by others from the reported results of intracranial surgery and of conservative therapy based upon the Cooperative Study of Intracranial Aneurysms and Subarachnoid Hemorrhage4,7 require specific comment. Based upon these data, some authors have suggested that there is “an absence of convincing proof that surgery is an improvement over competent nonsurgical care.”6 One statement, which has been widely quoted, states that “30 to 33% mortality for intracranial surgery can be used as a guideline by the average neurosurgeon.”7 We cannot agree that this is an acceptable figure. The risk in the individual cases may be much lower than this, and with proper selection the average will also be lower.

Furthermore, our experience with repeated bleeding from aneurysms suggests that this hazard may have been underrated. Table 9 presents our experience with rebleeding. These figures do not reflect a “natural course,” since our policy is to operate as soon as possible. They are best considered as reflecting a minimal incidence of rebleeding. We believe that these patients provide some evaluation of nonoperative management since the risk of rebleeding would be the same whether surgery had been planned or not. The future intentions of the attending surgeon should not influence the course of the disorder.

TABLE 9

Incidence of rebleeding

No. of Cases%No. of Hemorrhages*
245/27589Bled at least once
116/24547Bled more than once
73/24530Bled two times
35/24514Bled three times
8/2453Bled four or more times

44 cases bled after more than 3 weeks; 14 cases bled after more than 3 months.

Since most patients eventually reach the optimal-risk category, if they survive, and since the mortality in this group is so low, there seems little reason to deny them the assurance that definitive obliteration of the aneurysm can provide. Nevertheless, the Cooperative Study of Intracranial Aneurysms and Subarachnoid Hemorrhage has shown that the danger of rebleeding is less than many of us have assumed and, clearly, many patients survive for long periods without treatment.4 The surgeon cannot, therefore, consider that he is dealing with an inevitably lethal lesion in which any salvage, however sparse, justifies his efforts.

Summary

We have reviewed 275 cases of intracranial aneurysm to determine those factors that are significantly related to surgical risk. We have discussed and emphasized the following conclusions:

  1. Intracranial aneurysmorrhaphy can be accomplished with an extremely low mortality rate in the absence of severe meningeal reaction, neurological deficit, or serious associated disease.

  2. Meningeal reaction alone increases surgical hazard.

  3. Prompt surgical intervention is important for patients admitted in good condition.

  4. Conservative therapy until the patient's condition improves is advisable for the more seriously ill.

References

  • 1.

    BotterellE. H.LougheedW. M.ScottJ. W.VandewaterS. L. Hypothermia, and interruption of carotid, or carotid and vertebral circulation, in the surgical management of intracranial aneurysmsJ. Neurosurg.195613:142.BotterellLougheedScottVandewaterJ. Neurosurg.13:1–42.

  • 2.

    HamiltonJ. G.FalconerM. A. Immediate and late results of surgery in cases of saccular intracranial aneurysmsJ. Neurosurg.195916 :514541.HamiltonFalconerJ. Neurosurg.16:514–541.

  • 3.

    HuntW. E.MeagherJ. N.BarnesJ. E. The management of intracranial aneurysmsJ. Neurosurg.196219:3440.HuntMeagherBarnesJ. Neurosurg.19:34–40.

  • 4.

    NishiokaH. Report on the cooperative study of intracranial aneurysms and subarachnoid hemorrhage. VII, Pt. 1. Evaluation of the conservative management of ruptured intracranial aneurysms. J. Neurosurg.196625:574592.NishiokaJ. Neurosurg.25:574–592.

  • 5.

    PoolJ. L.PottsD. G. Aneurysms and arteriovenous anomalies of the brain. Diagnosis and treatment.New York, N. Y.: Harper & Row1965463 pp.PoolPottsAneurysms and arteriovenous anomalies of the brain. Diagnosis and treatment.

  • 6.

    ScheinbergP. Spontaneous subarachnoid hemorrhage: What should be the treatment?Medical News19671:6. ScheinbergMedical News1:6.

  • 7.

    SkultetyF. M.NishiokaH. Report on the cooperative study of intracranial aneurysms and subarachnoid hemorrhage. VIII, Pt. 2. The results of intracranial surgery in the treatment of aneurysms. J. Neurosurg.196625:683704.SkultetyNishiokaJ. Neurosurg.25:683–704.

Discussion

Dr. Charles G. Drake: It has been a pleasure to have the opportunity to look over this paper and to be able to congratulate the authors on their delightful figures in good risk cases; one death in 71 asymptomatic patients is notable and is something we all can shoot for.

As I see it, Dr. Hunt makes four points:

  1. Definitive operation on patients who have shown little or no effect from bleeding carries little risk.

  2. Over 50% of those patients who are stuporous with neurological defects will recover after a judicious waiting period and become satisfactory risks for operation.

  3. In spite of what some would have us believe, rebleeding is still a serious hazard, even after 3 weeks.

  4. Just how misleading is the widely quoted figure that a mortality of 30 to 33% is what the average neurosurgeon should expect in the surgical treatment of aneurysms?

To dispose of the last point first, one only has to look at the results of operation according to the grade of the patient. Our own mortality figure of 25% for intracranial operation for ruptured aneurysm in 226 cases is not far short of that oft-quoted figure. However, it is obvious that the poor results occur in patients whose brains have been seriously hurt by the hemorrhage and who are sick and dying. A number harbored life-threatening clots which had to be removed. As a result of these experiences, all of us have learned to delay operation in Grade III and IV cases (except with significant clot) and that all Grade V cases die in any event; for we have not yet learned to obliterate an aneurysm early after bleeding without hurting a sick, swollen brain further. Nevertheless, the waiting period permitted 55% of Grade III patients to improve to a normal state of alertness, such that they were then excellent candidates for operation.

What is of significance is that the surgical mortality is very low in patients who are well after a hemorrhage; one death in 71 (1.5%) in this report, although we lost four of 82 such cases (5%). Even so, it is this sort of figure that the conservatives should be quoting because they can't come close to matching it, and their patients who survive have no real security, for the threat of rebleeding lingers for life, albeit in diminishing probability.

It is difficult to compare Dr. Hunt's figures with ours, for his modification of the Botterell classification puts his Grade II patients into our Group I, since although they had headache and a stiff neck, they were alert. Our drowsy Group II patients would fall into his Group III. You will note that once a patient is drowsy and a little confused, the mortality jumps, to 25% in our 52 Grade II patients; but as will be seen, this is largely the result of operating within a week. Although many of their cases were operated upon within 24 hours of admission, only 20 of the 71 asymptomatic patients were operated on in 7 days or less after bleeding.

We must be careful in presenting figures such as these to indicate the timing of the operation. When operation is delayed for 2 or 3 weeks, one can do almost anything to an aneurysm and its vessels then and get away with it. Another concern is the loss of lives occurring from rebleeding during the waiting period; we must keep this interval as short as possible. Dr. Hunt and his group have had an unusual number of patients who have had what Gillingham calls “a warning leak,” a mere leak of blood from the sac and sudden headache whose significance often is not recognized by the patient, his relatives, and even the physician. The story of minor initial bleeding from an aneurysm is quite common in 50% of 87 cases analyzed by Gillingham. It foretells that frank rupture is almost certain to occur in a few hours, days, or even weeks. Early operation is so safe in these patients that Gillingham has said they should be led protesting to the operating room solely because of a few red blood cells in their spinal fluid. What can be more important than disseminating this knowledge about the warning leak to our physicians who see the patient first? A harmless lumbar puncture will reveal the grim nature of a sudden unexplained headache.

I still think it is dangerous to operate within a week after a brisk hemorrhage, when there is severe headache, a stiff neck, and especially if the patient is drowsy, for there is a great danger of producing postoperative arterial spasm. The mortality in drowsy patients operated upon in less than 7 days was 35%; after a week it was only 17%.

Similarly, only two deaths occurred in 66 Grade I patients after a week (3%) as compared to 2 in 16 (12.5%) under 7 days. Arterial spasm appears not to occur with any degree of severity after 7 days. This is why we adopted the policy of waiting 7 days in good risk patients who had had a brisk hemorrhage (alert Grade I and drowsy Grade II). Of a total of 61 such selected patients, 7 died of rebleeding during the week, a mortality of delay of 11% or about one case in 10. However, the operative results in the 54 who survived the week were worthwhile, for there were only 2 deaths and 2 poor results.

The quality of survival, which is perhaps more important than the mortality figures, must be good. Here 42 were excellent results and 8 good, so that 50 of 54 cases were able to return to their normal living, a success rate of 93%. I question that this success rate or mortality, including the 7 deaths from rebleeding during the week of delay, could have been reached by earlier operation, at least from my experience with arterial spasm, swelling, and infarction. These figures have been shown to complement those of Dr. Hunt's, who has demonstrated so graphically how very safe operation can be in the alert patient without symptoms.

It is probably true that if we could learn how to keep a patient safe from rebleeding for a week, or longer in obtunded patients with cerebral symptoms, the problems of the surgery of ruptured intracranial aneurysms would nearly be solved.

Article Information

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Time between hemorrhage and surgery for Grade I, II, and III patients. The vertical bars represent the number of cases operated on each post-bleeding day (measured from the last known hemorrhage) for those cases that had bled. The grading was done at the time of operation. Note the number of cases operated within the first week, and within the first three weeks in each grade. Delay in surgery in Grades I and II beyond the first day or two was usually incurred prior to admission to our institutions.

  • View in gallery

    Classification of patients by grade at admission, at surgery for all surgical procedures, and at surgery for intracranial ligation alone. The mortality for each classification is shown as a black wedge.

References

1.

BotterellE. H.LougheedW. M.ScottJ. W.VandewaterS. L. Hypothermia, and interruption of carotid, or carotid and vertebral circulation, in the surgical management of intracranial aneurysmsJ. Neurosurg.195613:142.BotterellLougheedScottVandewaterJ. Neurosurg.13:1–42.

2.

HamiltonJ. G.FalconerM. A. Immediate and late results of surgery in cases of saccular intracranial aneurysmsJ. Neurosurg.195916 :514541.HamiltonFalconerJ. Neurosurg.16:514–541.

3.

HuntW. E.MeagherJ. N.BarnesJ. E. The management of intracranial aneurysmsJ. Neurosurg.196219:3440.HuntMeagherBarnesJ. Neurosurg.19:34–40.

4.

NishiokaH. Report on the cooperative study of intracranial aneurysms and subarachnoid hemorrhage. VII, Pt. 1. Evaluation of the conservative management of ruptured intracranial aneurysms. J. Neurosurg.196625:574592.NishiokaJ. Neurosurg.25:574–592.

5.

PoolJ. L.PottsD. G. Aneurysms and arteriovenous anomalies of the brain. Diagnosis and treatment.New York, N. Y.: Harper & Row1965463 pp.PoolPottsAneurysms and arteriovenous anomalies of the brain. Diagnosis and treatment.

6.

ScheinbergP. Spontaneous subarachnoid hemorrhage: What should be the treatment?Medical News19671:6. ScheinbergMedical News1:6.

7.

SkultetyF. M.NishiokaH. Report on the cooperative study of intracranial aneurysms and subarachnoid hemorrhage. VIII, Pt. 2. The results of intracranial surgery in the treatment of aneurysms. J. Neurosurg.196625:683704.SkultetyNishiokaJ. Neurosurg.25:683–704.

TrendMD

Cited By

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 60 60 60
Full Text Views 1078 1078 647
PDF Downloads 273 273 144
EPUB Downloads 0 0 0

PubMed

Google Scholar