R.D. Bharath*, M.K. Vasudev, P.N. Jayakumar**, G. Goel, J. Mek, S. Ravishankar***, K. Thennarasu**** 	

Department of Neuroimaging and Interventional Radiology, National Institute of Mental Health and Neurosciences; Bangalore, India ** Department of Radiology, Khoula Hospital; Muscat, Oman *** Department of RadioDiagnosis, BGS Global Hospital; Bangalore, India **** Department of Biostatistics, National Institute of Mental Health and Neurosciences; Bangalore, India

Key words: Serial DSA, Intracranial aneurysm

* Corresponding author: Dr Rose Dawn Bharath - DNB RadioDiagnosis, DM Neuroradiology - Assistant Professor - Department of Neuro Imaging and Interventional Radiology - National Institute of Mental Health and Neurosciences - Dr H Marigowda Road - Bangalore. 560029 - Tel.: 080 26995427 - 080 26995683 - E-mail: drrosedawn@yahoo.com

Introduction

Management of an unruptured aneurysm remains controversial ¹. There are a number of factors that determine whether patients should receive conservative management with observation or intervention by surgical clipping or endovascular coil occlusion (class 2b, level of evidence C)¹. Over the past decade preventive endovascular treatment is increasingly being considered for intracranial aneurysms due to the poor prognosis associated with subarachnoid hemorrhage. However, the ideal treatment would be to treat only those aneurysms which are likely to rupture when we also consider the risks of treatment in a patient with an unruptured aneurysm. Many studies using in vitro models, idealized computational models and from clinical experience have identified arterial hemodynamics, arterial wall biodynamics and mechanobiology as some of the major factors which cause an aneurysm to form and then to rupture. There are very few studies dealing with the in vivo characteristics of intracranial aneurysms in the pre or post rupture period. What prompted this study was our own observation of varying morphology of some aneurysms on follow-up. At our institution, like elsewhere, there is some delay between the diagnosis of an aneurysm and its treatment. During this period we found that some aneurysms which were found on the initial angiogram disappeared, leaving us with a dilemma as what to treat! Sometimes we have encountered aneurysms on follow-up of an angionegative SAH. Hence we conducted this study with an aim to understand the morphologic and flow-related findings in aneurysms. We compared 22 ruptured intracranial aneurysms with 19 unruptured aneurysms.

Materials and Methods

Seventy-nine patients referred for aneurysm coiling were evaluated retrospectively to identify those who underwent angiograms once at the time of diagnosis and second at the time of intervention, between July 1999 and December 2005. Eighty-eight angiograms in 37 patients were evaluated to assess 41 aneurysms for their alterations in location, size, shape, intraluminal thrombosis, and stasis of contrast within the aneurysm and presence of spasm in the adjacent vessels. Five patients had two aneurysms and two patients had more than two angiograms prior to their treatment. Clinical and imaging parameters of these patients were also correlated. Based on the clinical presentation there were two subgroups of patients, Group A patients presenting acutely with subarachnoid haemorrhage (SAH) and Group B patients who had no clinical or imaging features suggestive of bleed. Moss Group B patients presented with mass effect due to giant aneurysms.
All angiograms were acquired on a Biplane angiography suite. Townes’ and lateral projections were acquired in all patients and 3D rotational angiogram in a few patients. Volumes of aneurysms were calculated using the formula (d₁+d₂+d₃)³x0.02. Changes in size (CS) were calculated using the formula=volume of aneurysm at initial angiogram (Va₁)-volume of aneurysm at second angiogram (Va₂). Ratio of change in size (RCS) were calculated using the formula = volume of aneurysm at initial angiogram (Va₁) - volume of aneurysm at second angiogram (Va₂) / volume of aneurysm at initial angiogram (Va₁). Rate of change of size (rCS) was calculated using the formula change in size/ number of days between the angiograms. Test of normality was done and it was found that all values were distributed along a non Gaussian pattern. Hence non parametric analysis using Wilcoxon-Mann-Whitney ‘U’ count was performed for statistical analysis.

Results

Nineteen men and eighteen women were evaluated in the age group of 28-70 years. Among the 41 aneurysms evaluated 22 aneurysms belonged to the Group A category presenting with SAH and 19 aneurysms were in Group B presenting as aneurysms with no clinical or imaging feature of SAH. ICA aneurysms formed the major category in both groups numbering ten in Group A and seven in Group B. Relatively more basilar aneurysms were noted in Group B (5 Vs 3). The locations of the aneurysms are listed in table 1.

Table 1 Location of aneurysms. + Comparison of Group A with Group B

Comparing the initial size of aneurysms, those in Group A (median=1.04 mm³) were significantly (p= 0.010) smaller than lesions in Group B (median= 4.53 mm³). Aneurysms in Group A showed an increase in size with a median value of 0.12 mm³ and those in Group B showed a decrease in size with a median of 1.73 mm³ (p=0.019).The rCS was significant between the groups (p=0.017), with the Group A aneurysms showing increase in size with a median rate of 0.041 mm³/day and Group B showing a decrease in size with a median rate of 0.061 mm³/day (figure 1).

A

Figure 1 A 45-year-old normotensive patient presented with a history of subarachnoid bleed. DSA Towne’s view of left ICA Injection (A) reveals no evidence of aneurysm. ICA injection with Cross compression also did not reveal any aneurysm (B). Six weeks later, left carotid angiogram in Towne’s view (C) reveals a large ACom artery aneurysm projecting inferomedially.

Data of hypertensive status were available in 30 of the 37 patients. Among these 17 were in Group A and 13 were in Group B. Normotensive patients in Group A had significantly smaller (P=.005) aneurysms in the initial angiogram and they showed a trend towards an increase in size and growth rate in the follow-up angiogram. Hypertensive patients in both groups showed a tendency for a decrease in the size of the aneurysms with a rapid rate of change in size in the unruptured group with a median of 3.83 mm³/day.

Presence of stasis was noted in 13 out of 37 aneurysms in the initial angiogram. In three patients the aneurysm was not seen in the initial angiogram and in one patient capillary phase was not available to describe stasis accurately. Presence of stasis had no significant relation to the change of aneurysm in group A. Unruptured aneurysms in Group B revealed that stasis was significantly associated with larger aneurysms. These aneurysms having stasis at the time of initial angiogram had significantly reduced in size on follow-up (p=0.013) at a faster rate (p=0.017) (figure 2).

Spasm of adjacent vessels was noted in five of the 41 initial angiograms. Presence of spasm was associated with increase in size of aneurysm on follow-up in both Group A and Group B. In the presence of spasm ruptured aneurysms showed a significant increase in the (P=0.023) ratio of change in size. Group B patients with spasm had significantly larger aneurysms on the initial angiogram compared with patients without spasm (figure 3).

Discussion

Soon after Moniz introduced cerebral angiography in the early 1930s, it was discovered that ruptured intracranial aneurysms could rebleed. Surgeons felt compelled to treat these aneurysms to prevent this complication. During the pioneering times, most surgeons favored an aggressive, early operative approach to ruptured intracranial aneurysms. This early approach, however, was characterized by prohibitively high morbidity and mortality ^2,3. Subsequently it became commonplace for surgeons to delay operation for two to three weeks after the initial hemorrhage ⁴. However surgeons noticed that many potentially treatable patients died or were disabled while awaiting delayed surgery . With the advances in neuroanesthesia and brain relaxation techniques and the development of microsurgical techniques supplemented with aggressive perioperative care, surgeons were enticed to reconsider early surgery. Following the publication of the results from ICSTAS (International Cooperative study on the timing of Aneurysm Surgery) in 1990, early surgery became the preferred approach to ruptured aneurysms ^5,6. The advances in endovascular strategies witnessed in the last decade have rekindled the controversy surrounding the timing of treatment for aneurysms . Over the past decade preventive endovascular treatment is increasingly being considered for even unruptured intracranial aneurysms due to the poor prognosis associated with subarachnoid hemorrhage. However, the ideal treatment would be to treat only those aneurysms which are likely to rupture when we also consider the risks of treatment in a patient with an unruptured aneurysm. Interventional neuroradiologists and neurosurgeons are in desperate need of fundamental understanding of vascular pathophysiology to guide the treatment of anuerysms. Many studies using in vitro models, idealized computational models and from clinical experience have identified arterial hemodynamics, arterial wall biodynamics and mechanobiology as some of the major factors which cause an aneurysm to form and then to rupture. There are very few studies dealing with the in vivo characteristics of intracranial aneurysms in the pre or post rupture period. This pilot study was undertaken to understand the spectrum of changes occurring in intracranial aneurysms with the aim to understand the factors which make a ruptured aneurysm different from an unruptured one using Serial DSA.⁷⁴
In the ruptured aneurysm group, the aneurysms were small in size with a median size of 1.04 mm. This finding is in agreement with the literature stating that small aneurysms tend to rupture earlier . There was a tendency for these aneurysms to increase in size with the rate equal to 0.061mm3/day. The rate of change in size of these aneurysms was significantly rapid in the ruptured group in comparison with the unruptured group. Hence a rapid change in size in a small aneurysm could be considered a predictor for early rupture.⁸³
The increase in size of aneurysms were not related to hypertensive status of the patient as also reported by Seppo Juvela et Al . Normotensive patients in both the groups had smaller aneurysms with a median value of 0.26 mm which showed a higher tendency to increase in size with a rate equal to 0.026 mm³/day as against the hypertensive group who had a median rate of 0.0001 mm/day. This could mean that there are factors apart from hypertension that affect the rupture and the subsequent growth of these aneurysms. It has been found that in patients with myocardial infarction, angiotensin converting enzyme inhibition (ACE inhibitors) suppresses the temporal increase in infarct zone collagen and attenuates infarct zone expansion, thinning and bulging, and LV enlargement and aneurysm formation during healing after MI . This action of ACE inhibitors can also probably be extended to intracranial aneurysms. Until further data emerge we can hypothesize that the lack of significant increase in the size of aneurysm in hypertensive patients could reflect the protective effect some of the antihypertensive drugs have on remodeling of aneurysms.
When there was evidence of spasm in the adjacent vessels there was a significant increase in ratio of change in size in the ruptured aneurysm group. The causes for the increase in size can be due to endothelial injury secondary to ischemia causing thinning of the wall and laxity , altered hemodynamics at the neck causing altered force on the wall , or due to poor contrast filling causing apparent reduction in initial size of the aneurysm. Current evidence does not conclusively support one explanation over the others, and further work will be needed to address this issue.
Though the presence of stasis was associated with a decrease in the size of aneurysm, this did not reach statistical significance because most of the aneurysms in the ruptured group did not show evidence of stasis (figure 4). However stasis was significantly associated with a decrease in size of aneurysms in the unruptured group. This can be explained by the Froude Number . The Froude number is the ratio of flow inertia to the gravitational force. A lower Froude number is associated with sluggish flow and more settling of contrast. Thus stasis of contrast can be taken as an indirect evidence for sluggish flow and these aneurysms thus can be expected to reduce in size.
The main limitation of this study is related to its retrospective nature. We adopted the assessment of aneurysm volume to measure the size of the aneurysms. Data from literature measure aneurysms in a single largest dimension; hence the values found in this study could not be correlated with the other studies. We could not assess whether ACE inhibitor was the antihypertensive drug used for control of hypertension from the records, hence the hypothesis of a remodeling effect in intracranial aneurysms needs further evaluation.
In conclusion, we found significant differences between ruptured aneurysms and unruptured ones on serial DSA. Ruptured aneurysms are small and show a rapid increase in size. Hypertensive patients on medication tend to have aneurysms with reduced growth rates. The presence of spasm is associated with an increase in the size of the aneurysm in the post rupture period. When there is evidence of stasis, which is commonly seen in large unruptured aneurysms, there is a significant tendency for reduction in the size of aneurysms. ^{10,12,12,11,33,9}

References

1. AHA/ASA Issues Guidelines on Intracranial Neurointerventional Procedures CME News Author: Susan Jeffrey , CME Author: Hien T. Nghiem, MD.

2. Graf CJ: Results of direct attack on non fistulous intracranial aneurysms with remark on statistics. J Neurosurg 12: 146-153, 1955.

3. Pool JL: Timing and techniques in the intracranial surgery of ruptured aneurysms of the anterior communicating artery. J Neurosurg 19: 378-388, 1962.

4. Dumont AS, Lanzino G, Kassell NF: The ruptured aneurysm: Timing of occlusion IN Peter D. Le Roux, H. Richard Winn, David W. Newell; Management of cerebral aneurysms, Elsevier Inc: 22: 335-346, 2004.

5. Kassel NF, Torner JC, Haley EC Jr et Al: The International cooperative study on the timing of aneurysm surgery. 1: Overall management results. J Neurosurg 73: 18-36, 1990.

6. Kassel NF, Torner JC, Haley EC Jr et Al: The International cooperative study on the timing of aneurysm surgery. II: Surgical results. J Neurosurg 73: 37-47, 1990.

7. Brilstra EH, Rinkel GJE, van der Graff Y et Al: Treatment of intracranial aneurysms by embolization with coils. A systematic review. Stroke 30: 470-476, 1999.

8. Locksley HB: Natural history of subarachnoid hemorrhage, intracranial aneurysms and arteriovenous malformations. J Neurosurg 25: 321-368,1966.

9. Juvela S, Poussa K, Porras, M: Factors affecting formation and growth of intracranial aneurysms. A long-term follow-up study. Stroke 32: 485-491, 2001.

10. Wang Z-J, Hoffmann KR, Zhou W et Al: Contrast settling in cerebral aneurysm angiography. Physics Med Biol 50: 13; 3171-3181, 2005.

11. Jugdutt BI, Lucas A, Khan MI: Effect of angiotensin-converting enzyme inhibition on infarct collagen deposition and remodelling during healing after transmural canine myocardial infarction. Can J Cardiol 13 (7): 657-68, 1997.

12. Wakhloo AK, Baruch BL et Al: Flow dynamics in aneurysms. In: Peter D Le Roux (ed), Management of cerebral aneurysms. Saunders 2004: 99-117.