Introduction

   Much clinical and laboratory studies have been made with regard to the nature, pathophysiology and management of cerebral vasospasm, but the cerebral ischemia related to vasospasm remains as an important cause of morbidity and mortality in aneurysmal subarachnoid hemorrhage (SAH).³⁾⁹⁾ Over the past decades, all neurosurgeons have become familiar with the problem of cerebral vasospasm and with intravascular volume expansion with or without induced arterial hypertension in the postoperative management of ruptured intracranial aneurysm. Many reports described the resolutions of delayed ischemic neurologic deficits caused by vasospasm after volume expansion and/or hypertension, and they recommended the use of hypervolemic hypertensive (HH) therapy for prevention and management of postoperative vasospasm in patients with ruptured aneurysm.^1)2)3)8)10)
   In recent studies, the overall frequency of multiple intracranial aneurysms among patients with aneurysmal SAH was approximately between 18%and 34%.^4)7)13)14) Not infrequently, if the additional unsecured, unruptured aneurysm (UUA) is on the contralateral side or is inaccessible during surgery of the ruptured aneurysm, the UUA is unprotected during the postoperative period and may be subject to increased hemodynamic stress from volume expansion and hypertension. In such circumstances, neurosurgeons are likely to be concerned about the possible risk of hemorrhage from UUAs and wonder whether HH therapy should be carried out or not. 
   We analyzed our experience from 45 patients with UUAs after treatment of uptured aneurysms in patients with multiple intracranial aneurysms to assess the risk of hemorrhage in these UUAs during HH therapy for cerebral vasospasm. 

Materials and ethods

1. Patients and data analysis
   The authors at Gil Medical Center, Gachon University of Medicine and Science treated 1052 patients with uptured and unruptured intracranial aneurysms by clip placement and/or coil packing from 1995 to 2004. Of 1052 patients, a total of 201 patients harbored multiple aneurysms. Among these 201 patients, 45 patients had UUAs after acute treatment of the ruptured aneurysms. We conducted retrospective analysis on these 45 patients. We divided 45 patients into two groups by the application of HH therapy;a hypervolemic hypertensive therapy group (HHG) and a non-hypertensive hypervolemic therapy group (NHHG). Data were collected on the location, size and number of UUAs at risk, mean arterial systolic blood pressure (SBP) and mean central venous pressure (CVP) during HH therapy (if they had), the duration of HH therapy (if they had), serial head computed tomography (CT), transcranial Doppler (TCD), preoperative and postoperative angiography, and neurological findings including the occurrence of symptomatic cerebral vasospasm. Then, we analyzed and compared the data between HHG and NHHG. The Pearson chi-square test and Student t test were used to compare the categorical or continuous variables between both groups. P<0. 05 was considered to represent a statistical significance. 

2. Patient management
   Surgical management of patients with ruptured intracranial aneurysm was clipping through craniotomy or coiling by the endovascular approach of the aneurysm that had been thought to be ruptured as well as any other accessible aneurysms during the same approach in the acute period after SAH. If the aneurysm suspected to be ruptured proved to be unruptured upon inspection, the remaining aneurysms were secured during the initial session of surgery or treated as soon as possible using another approach in the second stage. 
   Postoperatively, all patients with uptured aneurysms were monitored and managed in the intensive care unit (ICU) for about 2 weeks. Routine laboratory tests including blood count, electrolytes, urine, creatinine, blood urea nitrogen, liver enzymes and arterial blood gases were made at least once a day. A chest x-ray examination was performed daily and TCD was applied as needed. Routine head CT was taken at postoperative days 1, 7, and 14. Additional head CT was taken in the event of the onset of new neurologic deficit to rule out hemorrhage, mass effect, brain swelling, ischemic lesion and hydrocephalus. The SBP and CVP were measured hourly in ICU, and the 24-hour mean SBP and the 24-hour mean CVP were calculated from these hourly measurements. 
   Symptomatic vasospasm was defined as delayed neurologic deterioration that could not be attributed to rebleeding, hydrocephalus, intracerebral hematoma, seizure, electrolyte abnormalities or metabolic factors. Diagnosis of symptomatic vasospasm was determined by the findings of neurologic examination, cerebral angiography, laboratory tests and head CT supplemented with TCD. 
   During this study period, we routinely instituted prophylactic HH therapy immediately after securing the ruptured aneurysm and additional intra-arterial injection of papaverine and/or transluminal angioplasty by the endovascular approach for severe refractory vasospasm in patients with single aneurysm, but not in patients with multiple aneurysms where the UUAs remained. Instead, we did HH therapy without papaverine injection or angioplasty only when the patients retaining UUAs developed symptomatic vasospasm. 
   Volume expansion was induced by intravenous administration of crystalloid and colloid solutions such as normal saline and 5%albumin, and hypertension by dopamine and/or norepinephrine which were individualized to each patient depending on the cardiovascular esponse. The degree of induced hypervloemia and hypertension was adjusted to each patient in accordance with the severity of new neurologic deficits, TCD velocities, angiographic findings, and cardiovascular status. HH therapy was usually maintained until the resolution or stabilization of delayed ischemic neurologic deficit and then gradually discontinued over a few days under close observation. 

Results

1. Patient characteristics
   Twenty nine out of 45 patients who carried multiple intracranial aneurysms with UUAs, developed symptomatic vasospasm after treatment of ruptured aneurysms and had been managed with HH therapy. The remaining 16 patients, who did not show symptomatic vasospasm after initial surgery, had not been treated with volume expansion or hypertension. 
   The mean age, male to female ratio, Hunt-Hess grade and Fisher 's grade at first presentation of these 45 patients are summarized in Table 1. As expected, initial neurologic status and head CT grade were worse in HHG than NHHG, but there was no statistically significant difference in baseline characteristics between HHG and NHHG except male to female ratio as detailed on Table 1. 

2. Aneurysm characteristics
   All 45 patients were known preoperatively to have multiple aneurysms and the total number of aneurysms for these 45 patients was 103. Among the 103 aneurysms, 53 aneurysms were treated at initial surgery. As a result, 50 aneurysms were unprotected and at risk of hemorrhage, 30 of which were included in HHG and thus exposed to HH therapy. 
   All the UUAs except one were smaller than 10 mm in diameter. The most common location of UUAs was the middle cerebral artery contralateral to the ruptured aneurysm in both HHG (50. 0%) and NHHG (45. 0%). The numbers, sizes and locations of UUAs in both groups are listed in Table 2. 

3. Hemodynamic measurements and risk of hemorrhage
   In the 29 patients (HHG) with symptomatic vasospasm, the mean duration of HH therapy was 10.2 ±5.1 days (range, 4~20 days). The differences in SBP and CVP between HHG and NHHG were statistically significant as shown in Table 3. 
   No patient in both HHG and NHHG suffered hemorrhage from UUAs, even during HH therapy. Also, serial head CTs revealed no evidence of additional hemorrhage in all 45 patients. 
   Three of 29 patients in the HHG died in which 2 patients were afflicted by the massive cerebral infarctions from cerebral vasospasm in spite of HH therapy at postoperative day 17 and 19, respectively, and 1 patient suffered acute myocardial infarction at postoperative day 48 during rehabilitation. Repeated head CTs disclosed no new hemorrhage in these 3 patients. Of the remaining 42 patients, 34 patients had been treated on the UUAs by clipping or coiling at the period between 10 days and 84 months after initial treatment of ruptured aneurysms. Five patients in HHG and 3 patients in NHHG had not been operated on for various reasons including fixed severe neurologic deficits, and they had been followed up at the outpatient clinic for 4.2 to 72.5 months (mean 27.1 months in total, 17.4 months in HHG, and 43.4 months in NHHG)without an event of emorrhage from their UUAs. 

Discussion

   It is not uncommon for neurosurgeons to be confronted with multiple intracranial aneurysms and it may not be possible, for many reasons, to secure all of these aneurysms at the time when the ruptured aneurysm is treated since the reported incidence of multiple intracranial aneurysms is approximately between 18%and 34%.^4)7)13)14) Then, neurosurgeons may often encounter circumstances where a patient who sustains an aneurysmal SAH through surgery for the ruptured aneurysm bears other UUAs. If this patient develops symptomatic cerebral vasospasm requiring HH therapy, neurosurgeons may hesitate whether to induce hypervolemia and/or hypertension or not.
   Despite emarkable advancement in microneurosurgery, pharmacologic agents, and monitoring in ICU, cerebral vasospasm is a major cause of morbidity and mortality after aneurysmal SAH.³⁾⁹⁾ In 1976, management of aneurysmal SAH related symptomatic vasospasm with HH therapy was introduced by Kosnik and Hunt.¹⁰⁾ Since then, volume expansion with or without hypertension has been verified to be effective in eversing symptomatic vasospasm, and the use of HH therapy has become widely accepted as a prophylactic measure or as the treatment of choice for cerebral vasospasm in patients with ruptured aneurysms.^1)2)3)8)
   Hypertension is considered to be a risk factor for rebleeding in unsecured, ruptured aneurysms.⁸⁾¹²⁾ However, the burden of hypertension and/or hypervolemia upon the risk of hemorrhage in unruptured aneurysm is not evident.^6)11)15) A report by Kassell et al⁸⁾ demonstrated the efficacy of HH therapy in aneurysmal SAH related cerebral vasospasm as well as the negative effect of HH therapy on the rebleeding of unsecured uptured aneurysms. Their study indicates 3 (19%) of 16 patients who had unsecured uptured aneurysms suffered recurrent bleeding under HH therapy. In 6 other patients who underwent treatment for ruptured aneurysms but harbored UUAs, there was no hemorrhage from UUAs during HH therapy. In addition, they pointed out a patient in another series who had a fatal rupture of a previously intact giant carotid aneurysm under HH therapy. 
   Swift and Solomon¹⁶⁾ investigated the influence of HH therapy on the possible risk of bleeding from UUAs in 31 patients with not only treated ruptured aneurysms but also unprotected, unruptured aneurysms. Following surgery of ruptured aneurysms, they managed 31 patients with 34 UUAs through a protocol of prophylactic volume expansion with or without induced hypertension in which the maximum SBP was 202 ±53 mm Hg and the maximum CVP was 14.9 ±3.2 mm Hg. The duration of HH therapy was approximately 4 to 12 days. None of the 31 patients suffered rupture of UUAs du ing HH the apy. Recently, Hoh et al⁵⁾ reported their retrospective review of 73 UUAs at risk in 40 patients with multiple aneurysms who had surgery for ruptured aneurysms and had been managed with HH therapy for cerebral vasospasm. The mean SBP, mean CVP and mean course of HH therapy were 181.41 ±16.64 mm Hg, 10. 43 ±3.89 mmHg, and 7.25 days, respectively. Moreover, transluminal angioplasty and/or intra-arterial papaverine injection by endovascular approach were added to HH therapy in 11 of the 40 patients. Even with this invasive management for cerebral vasospasm, there was no instance of hemorrhage from the 73 UUAs during HH therapy and follow-up period of 121.75 aneurysm-years of risk. 
   In our study of 45 patients with UUAs, there was no hemorrhage in NHHG as well as HHG although statistically significant differences in SBP and CVP were made between the two groups. This result and protocol of HH therapy correspond with the reports by Swift and Solomon,¹⁶⁾ and Hoh et al.⁵⁾ However, we did not use intra-arterial papaverine injection or transluminal angioplasty as in the report by Hoh et al.⁵⁾ With regard to the size of UUAs, our study contains only 1 (3.4%) large aneurysm (≥10 mm) out of 30 UUAs at risk, whereas the report by Swift and Solomon,¹⁶⁾ and Hoh et al⁵⁾ consisted of 10 (29.4%) large or giant aneurysms in 34 UUAs and 5 (6.8%) large aneurysms in 70 UUAs, respectively. 
   We had operated on 34 of 45 patients with UUAs in a period between 10 days and 84 months after the first treatment of ruptured aneurysms. At outpatient clinic, we had observed 8 patients (5 patients in HHG and 3 patients in NHHG) who had UUAs for a mean period of 27. 1 months, but no patient had experienced hemorrhage from the UUAs. Hoh et al⁵⁾ also followed up the patients with UUAs (28 of 73 UUAs) for a mean of 6. 93 months until the second surgery, there was no instance of hemorrhage. In the study by Swift and Solomon,¹⁶⁾ 24 out of 31 patients with UUAs ultimately underwent clipping of their remaining aneurysms and they made mention of a patient who suffered an SAH from UUA while an outpatient awaited surgery, but they did not describe the follow-up period of the patients with UUAs.
   There are several limitations in this study. Our study includes a relatively small number of patients in a single institution. Concerning the size of UUAs, all the unsecured aneurysms but 1 aneurysm in our study were small (<10 mm) in size. Therefore, the risk of hemorrhage from large or giant UUAs from HH therapy is not estimated, because aneurysm size is known to be one of the important factors in the rupture rates of unruptured aneurysms.⁶⁾ Our study results are also insufficient for the long-term effect of HH therapy on the UUAs since the hemodynamic stresses on the UUAs may modify the long-term risk of hemorrhage from these aneurysms as indicated by Hoh et al.⁵⁾
   Meanwhile, there may be some different views on the timing of intervention for the UUAs. That is, whether clipping or coiling of incidental aneurysms in the acute period after aneurysmal SAH is advisable or not. Even though it is beyond the scope of our study, we prefer to treat such aneurysms as scheduled surgery after the patients survive from the initial bleeding from the ruptured aneurysms. 
   In our institute, the principles of management for aneurysmal SAH in this study period were to treat a uptured aneurysm as soon as possible, mostly within 24 hours after the attack, either surgically or endovascularly and then, to initiate prophylactic HH therapy. However, we were nervous about the prophylactic HH therapy and endovascular approaches for symptomatic vasospasm in patients with multiple aneurysms who had UUAs that could not be treated simultaneously during the first craniotomy or endovascular intervention. The authors did HH therapy in selected cases of patients with UUAs (29 out of 45 patients) where symptomatic vasospasm developed. 

Conclusions

   The authors study supports the safety of HH therapy for cerebral vasospasm following treatment of a ruptured aneurysm in the presence of UUAs of small size in patients with multiple aneurysms. To establish the risks of hemorrhage from larger UUAs and to examine the long-term risk of hemorrhage from the HH stress on the untreated aneurysms, meta-analysis from multiple institutions is anticipated. 

REFERENCES

1. Awad IA, Carter LP, Spetzler RF, et al. Clinical vasospasm after subarachnoid hemorrhage:response to hypervolemic hemodilution and arterial hypertension. Stroke 18:365-72, 1987

2. Dorsch NWC. A review of cerebral vasospasm in aneurysmal subarachnoid haemorrhage. Part II:Management. J Clin Neurosci 1:78-92, 1994

3. Dorsch NWC. Cerebral arterial spasm a clinical review. Br J Neurosurg 9:403-12, 1995

4. Ellamushi HE, Grieve JP, Jager HR, et al. Risk factors for the formation of multiple intracranial aneurysms. J Neurosurg 94:728-32, 2001

5. Hoh BL, Carter BS, Ogilvy CS. Risk of hemorrhage from unsecured, unruptured aneurysms during and after hypertensive hypervolemic therapy. Neurosurgery 50:1207-12, 2002

6. Juvela S, Porras M, Poussa K. Natural history of unruptured intracranial aneurysms:probability of and risk factors for aneurysm rupture. J Neurosurg 93:379-87, 2000

7. Kaminogo M, Yonekura M, Shibata S. Incidence and outcome of multiple intracranial aneurysms in a defined population. Stroke 34:16-21, 2003

8. Kassell NF, Peerless SJ, Durward QJ, et al. Treatment of ischemic deficits from vasospasm with intravascular volume expansion and induced arterial hypertension. Neurosurgery 11:337-43, 1982

9. Kassell NF, Torner JC, Haley EC Jr, et al. The international cooperative study on the timing of aneurysm surgery. Part 1:Overall management results. J Neurosurg 73:18-36, 1990

10. Kosnik EJ, Hunt WE. Postoperative hypertension in the management of patients with intracranial arterial aneurysms. J Neurosurg 45:148-54, 1976

11. Ohashi Y, Horikoshi T, Sugita M, et al. Size of cerebral aneurysms and related factors in patients with subarachnoid hemorrhage. Surg Neurol 61:239-47, 2004

12. Ohkuma H, Tsurutani H, Suzuki S. Incidence and significance of early aneurysmal rebleeding before neurosurgical or neurological management. Stroke 32:1176-80, 2001

13. Qureshi AI, Suarez JI, Parekh PD, et al. Risk factors for multiple intracranial aneurysms. Neurosurgery 1998;43:22-6. 

14. Rinne J, Hernesniemi J, Puranen M, et al. Multiple intracranial aneurysms in a defined population:prospective angiographic and clinical study. Neurosurgery 35:803-8, 1994

15. Shojima M, Oshima M, Takagi K, et al. Role of the bloodstream impacting force and the local pressure elevation in the rupture of cerebral aneurysms. Stroke 36:1933-8, 2005

16. Swift DM, Solomon RA. Unruptured aneurysms and postoperative volume expansion. J Neurosurg 77:908-10, 1992