Effect of early operation for ruptured aneurysms on prevention of delayed ischemic symptoms

Transcription

1 J Neurosurg 57: , 1982 Effect of early operation for ruptured aneurysms on prevention of delayed ischemic symptoms MAMORU TANEDA, M.D. Department of Neurosurgery, Hanwa Memorial Hospital, Osaka, Japan ~/ The effect of removal of subarachnoid blood clots on the prevention of delayed ischemic deficit was evaluated in 239 consecutive patients with ruptured supratentorial non-giant aneurysms. All patients were hospitalized within 24 hours after subarachnoid hemorrhage (SAH) and were classified in Grades 1 to 4 according to the system of Hunt and Hess; classification was made immediately preoperatively in patients operated on within 48 hours after SAH, or 48 hours after SAH in patients for whom delayed operation was planned. Delayed ischemic deficit causing permanent disability or death occurred in 11 (25%) of 44 patients in whom surgery was planned to be delayed for 10 days or more, in 26 (27.7%) of 94 patients in whom the aneurysms were obliterated and blood clots adjacent to them were removed within 48 hours of SAH, and in 11 (10.9%) of 101 patients in whom the aneurysms were obliterated and extensive and aggressive removal of thick subarachnoid clots lying along the arteries (identified on computerized tomographic scan) was performed within 48 hours of SAH. Accordingly, early operation is an effective and reliable method to reduce the occurrence of severe delayed ischemic deficit only when subarachnoid blood clots are removed extensively and aggressively along the arteries within 48 hours of SAH. KEy WORDS 9 cerebral aneurysm 9 ischemic deficit 9 early operation 9 subaracbnoid hemorrhage 9 vasospasm C EREBRAL vasospasm as a cause of delayed ischemia and infarction a,5,14,17 is a complication that can cause serious neurological deficits or death after rupture of an aneurysm.l,2,1s Prevention of vasospasm might reduce morbidity and mortality in patients with ruptured aneurysms. Unfortunately, at the present time, there are no reliable methods to prevent or treat cerebral vasospasm. In addition, the precise pathogenesis is not well understood. It has been strongly suggested that the extravasated blood in the subarachnoid space is related to the occurrence of vasospasm. 4,12,22 Surgical removal of subarachnoid blood clots before the onset of vasospasm has been proposed as a promising method to prevent or minimize delayed ischemia due to vasospasm. 9a5,16 However, Ljunggren, et al, la reported that early operation with removal of subarachnoid clots did not eliminate the risk of delayed ischemic dysfunction. Furthermore, an increased tendency for development of delayed ischemia was reported in patients who underwent early surgery. 1,~9 Such disagreement might have resulted from differences in patient population and techniques for cleaning the subarachnoid spaces. The present study was carried out in order to assess the usefulness of removing subarachnoid clots as one of the preventive measures against delayed ischemia. Three different groups of similar patients were treated in different ways for subarachnoid blood clots, and the outcome was compared. Patient Population Clinical Material and Methods Patients were selected for this study from 434 cases with ruptured supratentorial non-giant aneurysms treated during the past 8 years. All patients were hospitalized within 24 hours of subarachnoid hemorrhage (SAH). This series included 239 consecutive patients who were examined by angiography within 24 hours of SAH, and classified neurologically in Grades 1 to 4 according to the system of Hunt and Hess. 8 Patients in whom delayed operation was planned were classified neurologically at 48 hours after SAH, and patients who underwent operation within 48 hours of SAH were classified immediately preoperatively. In the 239 cases (152 women and 87 men), the 622 J. Neurosurg. / Volume 57/November, 1982

2 Anti-ischemic effect of early aneurysm surgery TABLE 1 Age distribution in 239patients with ruptured aneurysms Cases Age (yrs) No. Percent ~ : FIG. 1. Computerized tomography scans. Left: Preoperative scan showing blood in the subarachnoid space on the day of rupture of an anterior communicating artery aneurysm. Right: Postoperative scan showing decreased blood in the subarachnoid space after removal of the clots via a right pterional approach. The white spot in the center of the brain is the aneurysm clip. ruptured aneurysms were located on the internal carotid artery (ICA) in 75 patients, middle cerebral artery (MCA) in 69 patients, and anterior cerebral artery (ACA) in 95 patients. Age distribution is shown in Table 1. Management of Subarachnoid Clots The patients were classified into three groups according to the method of management of the subarachnoid blood clots: in 44 patients in Group 1, obliteration of the aneurysm was delayed for 10 days or more. Therefore, subarachnoid blood clots surrounded the arteries for at least 9 days. In several cases, however, ventricular drainage helped to remove blood-stained cerebrospinal fluid. During the interval before surgery, the patients were treated with bed rest, sedatives, analgesics, and antifibrinolytic agents. Blood pressure and intracranial pressure were monitored as needed. In 94 patients in Group 2, aneurysms were obliterated within 48 hours of SAH. A unilateral pterional approach was used in all except three cases with aneurysms of the distal ACA. During surgery, subarachnoid blood clots lying around the aneurysms and in the cisterns adjacent to them were removed completely. In many cases, the proximal Sylvian fissure was opened to remove the clots more extensively. In 101 patients in Group 3, extensive and aggressive removal of subarachnoid clots was performed at the time of aneurysm obliteration within 48 hours of SAH. The arachnoid membrane was invariably opened as extensively as possible via a unilateral pterional approach to remove the clots around the arteries. The arteries around which the arachnoid membrane could be opened by way of this approach included both ICA's, both proximal segments of the ACA, both sphenoidal segments of the MCA, and the insular segment of the MCA on the side of approach. TABLE 2 Clinical grades of 239 patients with ruptured aneurysms* Patient Grades 1 & 2 Grade 3 Grade 4 Total Group No. % No. % No. % Cases Group Group Group * Clinical grading according to Hunt and Hess. 8 Differences among the three groups were not significant by chi-square analysis. For a description of treatment groups see text. The interpeduncular and pontine cisterns were also opened to remove as much of the subarachnoid clots as possible. Thus, the subarachnoid blood collection (Fig. 1 left) was satisfactorily, although not completely, removed via a unilateral pterional approach (Fig. 1 rig,~t). This procedure was performed in all cases in this group, even if the subarachnoid blood collection was not demonstrated on computerized tomography (CT), because a considerable blood collection was sometimes encountered at locations where blood was not visualized on CT scans, especially in the insular cistern. In 17 cases, CT scanning demonstrated a ttfick or clearly visible subarachnoid blood collection in the insular cistern on the side opposite to the operative approach and/or in the interhemispheric fissure (Fig. 2 upper); this blood collection was also removed via an additional contralateral pterional approach and/or an interhemispheric approach (Fig. 2 lower). The clinical condition of the patients in each group is shown in Table 2; there were no statistically significant differences among the three groups. All operations in this series were performed by the author using the surgical microscope. Evaluation of Quantity of Subarachnoid Blood The quantity of subarachnoid blood was evaluated by CT before and after surgery in all Group 3 patients. Blood den,;ity was classified into three grades: not visible, obscurely visible, and dearly visible. If CT did not demon,;trate a high-density area in the subarachnoid space, the blood was judged as "not visible." "Obscurely visible" blood was clearly visualized as a J. Neurosurg. / Volume57~ November,

3 TABLE 3 Incidence of delayed ischemic symptoms and quality of outcome in relation to the timing and methods of operation M. Taneda Ischemia Ischemia Not Outcomew Treatment Group Total Responsible for Responsible for & Neurological Cases Bad Outcome] Bad Outcome1: Good Bad Grade* No. % No. % No. % No. % Group Grades 1 & Grade Grade Group Grades 1 & Grade Grade Group Grades 1 & Grade Grade * Clinical grading according to Hunt and Hess) For a description of treatment groups see text. t Statistical significance: Group 1 vs. Group 2: X 2 = 0.11, not statistically different; Group 2 vs. Group 3: X 2 = 8.91, p < 0.01; Group 1 vs. Group 3: X 2 = 4.74, p < :~ Statistical significance: Group 1 vs. Group 2: X 2 = 3.87, p < 0.05; Group 2 vs. Group 3: X 2 = 1.33, not statistically different; Group 1 vs. Group 3: X 2 = 7.05, p < w Statistical significance: Group 1 vs. Group 2: X 2 = 7.08, p < 0.01; Group 2 vs. Group 3: X 2 = 8.38, p < 0.01; Group 1 vs. Group 3: X 2 = 24.40, p < high-density area on CT scan; however, the contour was obscure and the density was apparently lower than that of a clot. "Clearly visible" blood was a clot with relatively well defined contours. The degree of blood was determined at the location in the subarachnoid space with the highest density, regardless of the extent of the blood collection. FIG. 2. Computerized tomography scans. Upper: Preoperative scans showing blood in the subarachnoid space on the day of rupture of an anterior communicating artery aneurysm. Note the clearly visible blood clots in the interhemispheric fissure and the bilateral insular cisterns which are impossible to remove via a unilateral pterional approach. Lower: Postoperative scans showing decreased blood in the subarachnoid space after removal of the clots via interhemispheric and bilateral pterional approaches. Assessment of Delayed Ischemic Symptoms and Outcomes If clinical deterioration, which could be positively assessed in Grade 1 to 4 patients, occurred several days after SAH, carotid angiography was performed. When the arterial caliber was less than half that observed on the angiogram performed within 24 hours of SAH, and the distribution of vasospasm corresponded to the clinical symptoms, ischemia due to vasospasm was considered to be responsible for the clinical deterioration. Rebleeding, hydrocephalus, and extracranial factors that might cause clinical deterioration (such as disturbances of fluid and electrolyte balance) were ruled out. Outcomes were determined at 6 months after SAH. The patients who had returned to full previous activity with or without minor neurological deficit were classified as having a good outcome. The patients who were unable to return to their previous activity or had died were classified as having a bad outcome. Results Incidence of Delayed Ischemic Symptoms Of 239 patients in this series, 80 patients developed delayed ischemic symptoms 4 to 16 days after SAH. 624 J. Neurosurg. / Volume 57/November, 1982

4 Anti-ischemic effect of early aneurysm surgery TABLE 4 Incidence of delayed ischemic symptoms and quality of outcome in Group 3 patients in relation to preoperative quantity of subarachnoid blood* Ischemia Ischemia Not Outcome Quantity of Total Responsible for Responsible for Subarachnoid Cases Bad Outcome Bad Outcome Good Bad Blood No. Percent No. Percent No. Percent No. Percent not visible Grades 1 & Grade Grade obscurely visible Grades 1 & Grade Grade 4 5 I clearly visible Grades 1 & Grade Grade * Clinical grading according to Hunt and Hess. 8 The average interval between SAH and onset of ischemic symptoms was 8.3 _+ 2.0 (mean + standard deviation) days in Group 1, 7.2 _+ 2.1 days in Group 2, and 7.4 +_ 3.0 days in Group 3. There were no statisticauy significant differences among them. As shown in Table 3, the incidence of ischemic symptoms that were not responsible for a bad outcome (mild ischemia) increased significantly in Group 2 as compared with Group 1. The incidence of delayed ischemia that was responsible for a bad outcome (severe ischemia) was 27.7% (26 of 94 patients) in Group 2, as opposed to 25% (11 of 44 patients) in Group 1. There was no statistical significance between these groups. On the other hand, the incidence of severe ischemia in Group 3 patients who underwent extensive and aggressive removal of subarachnoid clots within 48 hours of SAH was 10.9% (11 of 101 patients), and this was significantly different from the incidence of severe ischemia in Group 1 or 2. The tendency to a decreased incidence of severe ischemia in Group 3 patients seems to be present in every clinical grade. The difference in the incidence of mild ischemia was not statistically significant between Groups 2 and 3. Timing of Operation, Removal of Blood Clots, and Outcome There were significant differences in outcome when the three treatment groups were compared (Table 3). The outcome was most favorable in Group 3 and most unfavorable in Group 1. The influence of timing of operation on the outcome was examined by comparing the outcomes between Groups 1 and 2. The outcome in Group 2 was significantly better than that in Group 1, in spite of failure to reduce the incidence of ischemic deficit. This may reflect prevention of recurrent hemorrhage which occurred later than 48 hours after SAH, and resulted in death or permanent morbidity in 18.2% (eight of 44) in Group 1. The effects of removal of subarachnoid clots on the outcome were assessed, comparing the results in Group 2 and 3 patients. The outcome in Group 3 was significantly better than that in Group 2. Effect of Quantity of Subarachnoid Blood in Group 3 The preoperative quantity of subarachnoid blood estimated by CT scan, clinical grades, incidence of delayed ischemic symptoms, and outcomes in Group 3 patients (in whom aggressive and extensive removal of subaracjhnoid clots was performed) are summarized in Table 4. There was no strict correlation between the quantity of blood and clinical grades, although CT scans of patients in neurological Grade 4 always demonstrated blood collection, and those of the more severe grades of the patients tended to demonstrate a larger amount of blood. The differences in incidence of ischemia responsible for bad outcome were not statistically significant among the three preoperative blood quantity groups. Correlation of the postoperative quantity of blood versus incidence of ischemia responsible for bad outcome was not made, because the satisfactory, although incomplete, removal of subarachnoid clots was confirmed by CT scans performed immediately postoperatively in all cases (as shown in Figs. 1 and 2), and there was no noticeable difference in the quantity of blood that remained postoperatively. Outcomes were most favorable in the group without visible blood collection, although a significant difference was noted only between the group without visible blood and the group with obscurely visible blood J. Neurosurg. / Volume 57 / November,

5 M. Taneda (p < 0.05). This might be attributed to the better clinical grades in the patients without visible blood rather than to the smaller accumulation of blood. Site of Infarction in Patients with Early Surgery Delayed ischemia that was severe enough to be responsible for bad outcome was clearly delineated as a large low-density area by CT scan, although CT findings of mild ischemia were usually negative, temporarily positive, or permanently positive as a small or obscure lesion. On the other hand, the degree of angiographic vasospasm did not always correlate with the size or severity of infarction. Moreover, severe vasospasm did not always cause ischemic symptoms. Accordingly, failure of prevention of severe ischemia in patients having early operation was examined by CT scan, which was most reliable in demonstrating morphologically the extent and location of ischemia. The results are shown in Table 5. In Group 2, 12 of 26 patients who developed severe ischemia were examined by CT scan. They had all been operated on via the unilateral pterional approach. Infarcted areas were demonstrated by CT on the side of the approach in seven cases, on both sides in two cases, and on the side opposite to the approach in three cases. Three ruptured aneurysms with infarcts on the side opposite to the approach were on the anterior communicating artery in two cases and the ICA on the side of the approach in one case. In Group 3, all of the 11 patients who developed severe ischemia were examined by CT scan. They had all been operated on via a unilateral pterional approach. Of the 17 patients in whom blood clots were removed via a unilateral pterional approach as well as by the contralateral pterional approach and/or interhemispheric approach, none developed severe ischemia. Infarction was demonstrated on CT scan on the side of the surgical approach in four cases, on both sides in two cases, and on the side opposite to TABLE 5 Site of infarction responsible for bad outcome in patients operated on early* Site of Clinical Grader Total Infarction 1 & Group ipsilateral to approach both sides contralateral to approach Group ipsilateral to approach both sides contralateral to approach * All of these patients were operated on via a unilateral pterional approach. Site of infarction was determined by computerized tomography. Group 3 patients had extensive removal of clots, whereas Group 2 patients did not. t Clinical grading according to Hunt and Hess. 8 the approach in five cases. In three of four cases with infarcts ipsilateral to the approach, the area of ischemia was in the territory of the MCA on the side of approach, where removal of clots was assumed to have been performed most aggressively and extensively. Five ruptured aneurysms with infarcts on the side opposite to the approach were located on the anterior communicating artery in two cases and the ICA on the side of the approach in three cases. Discussion Early surgical intervention for ruptured aneurysms has been suggested as a promising policy to achieve good results by preventing vasospasm as well as recurrent hemorrhage. 9,15,16,2~ Weir and Aronyk 21 reported the superiority of early over late operations only for Grade 3 and 4 patients. However, the factors by which early operation contributed to improvement of the results in such patients were not analyzed. Hugenholtz and Elgie 7 did not consider that early surgery prevented the appearance of vasospasm, although they concluded that it was a reasonable alternative to a policy of delayed operation in some cases. Ljunggren, et al., '1 failed to provide evidence that cerebral delayed ischemic dysfunction could be eliminated by operation performed within 48 hours of rupture, but they found early operation improved overall outcome by preventing recurrent bleeding. In the present study, the outcome in Group 2 and 3 patients who underwent early operation was significantly superior to that in Group 1 patients who were not operated on early. However, in Group 2 patients, the superiority was probably associated with prevention of rebleeding by clipping the aneurysms, and not by prevention of delayed ischemia. On the contrary, the incidence of delayed ischemic symptoms in Group 2 patients increased rather than decreased. These results do not exclude the role of subarachnoid blood in the etiology of vasospasm. Subarachnoid blood is widely dispersed, and removal of blood only around the aneurysms and from the cisterns adjacent to them would not be sufficient to prevent delayed ischemia. Although early operation did not prevent delayed ischemia in Group 2 patients, it significantly reduced the incidence of delayed ischemia that was severe enough to cause death or permanent disability in Group 3 patients. This was the most important factor in achieving a better result in Group 3 patients than in Group 2 patients. Such results indicate that insufficient removal of subarachnoid clots does not contribute to improvement in the results by reducing ischemia; the protective effect of early operation for delayed ischemia may be obtained only by extensive and aggressive removal of subarachnoid clots along the arteries. Vasospasm is usually more severe on the side of the brain ipsilateral to the aneurysms. 5,23 However, in this series, occurrence of infarction due to vasospasm on 626 J. Neurosurg. / Volume 57 / November, 1982

6 Anti-ischemic effect of early aneurysm surgery the side contralateral to the ruptured aneurysms was confirmed by CT scan in four patients operated on early. All of those aneurysms were located on the ICA, usually on the side from which the subarachnoid clots were removed. Moreover, infarction was apt to occur less frequently on the side where periarterial blood clots were removed extensively and aggressively, as observed in Group 3 patients (Table 5). It is, therefore, suggested that removal of subarachnoid clots is effective to prevent infarction due to vasospasm. In three Group 3 patients, severe ischemia occurred in the territory of the MCA, where satisfactory removal of clots was confirmed by postoperative CT scan. This might be explained as follows: First, there might be some etiological factors other than subarachnoid blood in the occurrence of vasospasm, although the blood would play a major role. Second, some pathological process leading to vasospasm might be immediately triggered by SAH, and its progression would not cease in some individuals despite removal of blood within 48 hours of SAH. It is quite possible that removal of clots later than 48 hours after SAH might check the process less effectively. The author has frequently encountered severe postoperative ischemia, in spite of extensive removal of dots, among patients operated on between the 4th and 7th days following a hemorrhage, as reported by others. 6,7,~6 This finding might also reflect an increased vasoconstrictive response to mechanical stimuli between the 4th and 7th days after SAH. Even within 48 hours of SAH, mechanical stimuli to the arteries might increase the incidence of delayed ischemia if subarachnoid clots were not radically removed, as observed in Group 2 patients. Thus, therapy should be aimed at prompt removal of subarachnoid clots to prevent vasospasm. Although extensive and aggressive removal of subarachnoid blood clots seemed to be necessary to prevent or minimize delayed ischemic deficits, it sometimes caused transient neurological deterioration, especially when the patients were very old. The brain in the acute phase was sometimes soft if the patient was in a grave condition, and manipulation of the brain must be as gentle as possible. Forceful brain retraction should be absolutely avoided. Among the three treatment groups, Group 3 achieved the best overall outcome. However, good recovery, in which the patients returned to full previous activity, was achieved in only 69.3% of patients: It was achieved in 91.7% of Grade 1 and 2 patients, in 69.6% of Grade 3 patients, and in 51.6% of Grade 4 patients. The operative results are less favorable than results reported previously; la,24,25 however, many patients in those reports seem to have been carefully selected or to have undergone delayed operation. Accordingly, it is difficult to make a comparison in the results between our Group 3 and those reports. Kassell, et al., ~~ reported the time from SAH to ad- mission to be an important parameter in predicting outcome, because patients admitted on Days 0 to 3 after SAH did worse than those admitted on Days 4 to 7, indicating that comparison should be made of groups adimitted at similar intervals following the ictus. The patient population in this series consisted of consecutive unselected patients, all of whom were hospitalized within 24 hours after SAH. Moreover, there were no statistically significant differences among the three treatment groups in relation to clinical condition. Hence, comparison among them is assumed to be strict. The results in this study indicate that early intracranial operation for a ruptured aneurysm is an effective and reliable method to reduce the occurrence of severe delayed ischemic deficit only when subarachnoid blood clots are removed extensively and aggressively along the arteries within 48 hours after SAH. References 1. Allcock JM, Drake CG: Postoperative angiography in cases of ruptured intracranial aneurysm. J Nenrosurg 20: , Artiola i Fortuny L, Prieto-Valiente L: Long-term prognosis in surgically treated intracranial aneurysms. Part 1: Mortality. J Neurosurg 54:26-34, Crompton MR: The pathogenesis of cerebral infarction following the rupture of cerebral berry aneurysms. Brain 87: , Fisher CM, Kistler JP, Davis JM: Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery 6:1-9, ] Fisher CM, Roberson GH, Ojemann RG: Cerebral vasospasm with ruptured saccular aneurysm--the clinical manifestations. Neurosurgery 1: , Hori S, Suzuki J: Early and late results of intracranial direct surgery of anterior communicating artery aneurysms. J Neurosurg 50: , Hugenlioltz H, Elgie RG: Considerations in early surgery oil good-risk patients with ruptured intracranial aneurysms. J Neurosurg 56: , Hunt WE, Hess RM: Surgical risk as related to time of intervention in the repair of intracranial aneurysms. J Neurosurg 28:14-20, Johnso~a RJ, Potter JM, Reid RG: Arterial spasm in subarachnoid hemorrhage: mechanical considerations. J Neurol Neurosurg Psychiatry 21:68, 1958 (Abstract) 10. Kassell NF, Adams HP Jr, Torner JC, et al: Influence of timing of admission after aneurysmal subarachnoid hemorrhage on overall outcome. Report of the Cooperative Aneurysm Study. Stroke 12: , Ljunggren B, Brandt L, Kfigstr6m E, et al: Results of early operations for ruptured aneurysms. J Neurosurg 54: , Mizukami M, Takemae T, Tazawa T, et al: Value of computed tomography in the prediction of cerebral vasospasm after aneurysm rupture. Seurosurgery 7: , Mullan S, Hanlon K, Brown F: Management of 136 consecutive supratentorial berry aneurysms. J Neurosurg 49: , 1978 J. Neurosurg. / Volume 57 / November,

7 M. Taneda 14. Odom GL: Cerebral vasospasm. Clin Neurosurg 22: 29-58, Pool JL: Early treatment of ruptured intracranial aneurysms of the circle of Willis with special clip technique. Bull NY Acad Med 35: , Saito I, Ueda Y, Sano K: Significance of vasospasm in the treatment of ruptured intracranial aneurysms. J Neurosurg 47:412429, Schneck SA, KrichefflI: Intracranial aneurysm rupture, vasospasm, and infarction. Arch Neurol 11: , Stornelli SA, French JD: Subarachnoid hemorrhage: factors in prognosis and management. J Neurosurg 21: , Sundt TM Jr, Whisnant JP: Subarachnoid hemorrhage from intracranial aneurysms. Surgical management and natural history of disease. N Engl J Med 299: , Suzuki J, Yoshimoto T: Early operation for the ruptured intracranial aneurysm. Jpn J Surg 3: , Weir B, Aronyk K: Management mortality and the timing of surgery for supratentorial aneurysms. J Neurosurg 54: , Wilkins RH: The role of intracranial arterial spasm in the timing of operations for aneurysms. Clin Neurosurg 24: , Wilkins RH, Alexander JA, Odom GL: Intracranial arterial spasm: a clinical analysis. J Neurosurg 29: , Yaw MG, Fox JL: The microsurgical approach to intracranial aneurysms. Surg Neurol 3:7-14, Yoshimoto T, Uchida K, Kaneko U, et al: An analysis of follow-up results of 1000 intracranial saccular aneurysms with definitive surgical treatment. J Neurosurg 50: , 1979 Manuscript received March 19, Accepted in final form June 18, Address reprint requests to: Mamoru Taneda, M.D., Department of Neurosurgery, Hanwa Memorial Hospital, 11-11, Karita-7, Sumiyoshi-ku, Osaka 558, Japan. 628 J. Neurosurg. / Volume 57 / November, 1982