Natural History of Unruptured Aneurysms

Intact or unruptured intracranial aneurysms (UIAs) are aneurysms with no recent or remote history of subarachnoid hemorrhage (SAH). Management of these patients is controversial because the natural history from the current studies is unclear.¹ The treating physician must evaluate not only the aneurysm but also patient factors such as age, medical condition, and family history. Such information allows the neurosurgeon to make a calculated judgment about the lifetime risk associated with UIAs in comparison to the risk related to treatment. For example, a relatively benign natural history would favor conservative therapy (i.e., observation), particularly in the older population. In contrast, a more malignant natural history in a younger patient would make intervention more urgent. Life tables, which are based on population studies, are an estimate of life expectancy (Fig. 360-1), but an individual may vary considerably from these averages. Therefore, assessment of an individual patient’s risk involves evaluating not only the likelihood of rupture of an intact aneurysm but also the patient’s life expectancy.

FIGURE 360-1 Life expectancy by age.

(From Mon Vit Stat Rep 1997;45:18.)

Prevalence of Unruptured Aneurysms

There is considerable variation in the reported prevalence of UIAs (Table 360-1). This variation can be attributed to multiple factors, such as the nature of the study (i.e., retrospective or prospective), the mechanism and accuracy of detection (i.e., autopsy, CT, computed tomographic angiography [CTA], MRI, magnetic resonance angiography [MRA], catheter angiography), the interest and focus of the investigator (i.e., pathologist, neuropathologist, or vascular neuropathologist), and definition of a cerebral aneurysm (i.e., 1-mm versus larger expansion at a vessel bifurcation). In addition, the characteristics of the population, such as age (e.g., elderly) and nationality (i.e., Finnish and Japanese), will influence the prevalence. Lifestyle (i.e., tobacco use) and environmental factors (alcohol) may also explain variations in reported prevalence rates of UIAs. Finally, it is becoming more evident that genetic influences are involved in aneurysm formation (for further discussion, please see Chapter 362).

TABLE 360-1 Prevalence of Unruptured Aneurysms

	NO. OF PATIENTS	PREVALENCE
Autopsy
Fox et al., 1983	164,764	0.8%
Radiology
Pre-CT era: Winn et al., 1978	4658	0.65%
Post-CT era: Ujie et al., 1993	1612	2.7%

CT, computed tomography.

Rinkel and colleagues comprehensively reviewed 23 studies including more than 50,000 patients and noted a prevalence, on average, of 0.4% in retrospective autopsy studies, 3.6% in prospective autopsy studies, 3.7% in retrospective angiography studies, and 6.0% in prospective angiography studies.² As just noted, this variation can be attributed to the nature of the population studied and the mechanisms used to document the presence of a UIA. Most of the angiographic studies analyzed by Rinkel and coworkers originated in the post-CT/MRI era, which may have skewed the prevalence of UIAs upward.² In the post-CT/MRI era, angiography is used mainly for patients with vascular disease and therefore encompasses an older segment of the patient population. In contrast, a pre-CT (see Table 360-1) study based on angiography revealed a prevalence of 0.65%. In the pre-CT era, angiography was used more liberally; a broader segment of the population was surveyed and therefore may be more representative of the true prevalence of UIAs.

In summary, the prevalence of UIAs has been found to vary considerably from less than 1% to as high as 9%. This variation appears to be related to a number of factors, such as methodologies and techniques used in data collection, environmental influences, patient characteristics, and genetic factors.

Rupture Rate of Unruptured Aneurysms

Despite intensive investigations, the incidence of SAH in patients with UIAs is not perfectly defined, as shown in the following studies (summarized in Table 360-2):

• Locksley monitored 34 patients with 34 UIAs for almost 4 years and noted that 9 patients suffered an SAH, for a yearly rupture rate of 7%.³ However, all the patients described in Locksley’s report had symptomatic aneurysms, which may have a high risk for hemorrhage (see the later section “Symptomatic Aneurysms”).

• Zacks and colleagues reported data from 10 patients with 12 incidental aneurysms. No patients suffered an SAH; however, 5 of the UIAs were smaller than 3 mm, 3 were 3 to 6 mm, and 1 was 10 mm in diameter.⁴ One of the patients died of a pulmonary embolus 2 months after treatment. Three of the 9 survivors underwent follow-up angiography 12 to 16 months later, and no change in appearance of the aneurysm was noted. These 9 patients were observed for a mean of 37 months (maximum, 90 months), during which no SAH occurred.

• Heiskanen presented a follow-up review of 61 patients with 129 unruptured aneurysms who had previously been treated for ruptured aneurysms and were observed for 10 years.⁵ During this period 7 patients bled from a previously unruptured aneurysm, for a rupture rate of approximately 1.1% per year.

• Wiebers and coworkers reported a selected group of 65 patients with 81 unruptured intracranial saccular aneurysms who did not undergo surgery.⁶ In 8 of the 65 patients, SAH subsequently developed as a result of rupture of an aneurysm over a mean follow-up interval of slightly more than 8 years, for an approximate rupture rate of 1%. This study was particularly important in that it was first to stratify the lesions by size and to assess the value of aneurysm size in the risk for rupture. Wiebers and colleagues reported a zero risk of rupture for aneurysms less than 10 mm in diameter and a risk of 1.7% per year for aneurysms larger than 10 mm.⁶ Their data also suggested that patients with multiple aneurysms had an increased risk for rupture.

• In 1983, Winn and coauthors reported the long-term (10-year) outcomes of patients with multiple aneurysms.⁷ All patients had a history of SAH and surgical treatment of ruptured aneurysm. The investigators evaluated the fate of the intact aneurysms and found approximately a 1% per year rate of hemorrhage.

• 1987, Wiebers and associates further reported their long-term follow-up of 130 patients with 161 unruptured aneurysms.⁸ This study also included follow-up of the patients from their earlier analysis and confirmed their previous observations: unruptured saccular aneurysms less than 10 mm in diameter have a very low probability of subsequent rupture, with no patients who had aneurysms less than 10 mm in diameter rebleeding. However, 15 of the 59 aneurysms larger than 10 mm ruptured, for a rupture rate of approximately 2% per year for that group.

• Easton and coworkers speculated on the natural history by assuming a prevalence of 5% (based on autopsy rates) and a population-based incidence of aneurysmal SAH of 10 per 100,000 persons per year.⁹ These investigators concluded that most aneurysms never rupture. However, their prevalence rate of 5% is high (see Table 360-1), which may have overrepresented the number of patients at risk and thereby skewed the risk for rupture downward.

• In 1993, Juvela and colleagues reported 181 aneurysms in 142 patients with a mean follow-up of 166.8 months.¹⁰ They found that the risk for rupture was 1.4% annually, a value based on the first 27 aneurysm ruptures during the follow-up period. If considering aneurysms rather than patients, the annual risk was 1.1% (27 events during 2434 aneurysm-years). This population was somewhat different from that of Wiebers and colleagues in that 92% of the patients had suffered a previous SAH from another aneurysm but was similar to the population studied by Winn and associates.⁷

• In 1993, Asari and Ohmoto reported on 54 patients with 72 unruptured cerebral aneurysms.¹¹ Twenty-two patients died during the observation period, which averaged 43.7 months. The 5-year survival rate was 56%. Aneurysms ruptured in 11 patients (20.4%), 10 of whom died without undergoing surgery. The annual bleeding rate was 1.9%, and the average size of the 11 ruptured aneurysms was 13.1 mm. In 4 patients, however, bleeding occurred in unruptured cerebral aneurysms 4 to 5 mm in size.¹¹

• Taylor and coworkers reported a large study involving 20,767 Medicare patients who were admitted to a hospital and in whom unruptured cerebral aneurysms were diagnosed.¹² The average age of these patients was 73.8 years, and 70% were women. In this study, 2648 patients were excluded from follow-up because of a concurrent diagnosis of SAH or in-hospital death, and therefore 18,119 patients were identified with the diagnosis of unruptured cerebral aneurysm. These patients were divided into two populations. Group 1 (7113 patients) was composed of patients with UIA as a primary diagnosis, whereas group 2 (11,066 patients) consisted of patients with UIA as a secondary diagnosis. After 2.5 years of follow-up, the risk for hemorrhage was 2% per year for the first group and 1.3% for the second group.¹²

• In 1997, Yasui and colleagues reported the results of a follow-up study of 303 unruptured aneurysms in 234 patients.¹³ Single aneurysms were present in 171 patients and multiple aneurysms in 63. The mean follow-up period was 75 months (range, 3 to 270 months). Of the 234 patients, 132 (56.4%) survived, 59 (25.2%) died of other diseases, 9 (3.8%) underwent surgery, and 34 (14.5%) bled from unruptured aneurysms, which was fatal in 18 of the patients. The annual rupture rate was 2.3%.¹³ The cumulative probability of rupture was significantly higher for patients with multiple aneurysms than for those with single aneurysms.

• The International Study of Unruptured Intracranial Aneurysms (ISUIA), a study consisting of several groups of patients with relevance to the natural history of UIAs, divided 1449 patients with 1937 UIAs into two groups: 727 patients (group 1) who had no previous history of SAH and 722 patients (group 2) who had a history of SAH.¹⁴ The mean duration of follow-up was 8.3 years, with a total of 12,023 patient-years of follow-up. Of the 1449 patients, 32 had documented aneurysmal ruptures. In group 1, the cumulative rate of rupture was 0.05% per year for aneurysms less than 10 mm and about 1% for those larger than 10 mm in diameter. Aneurysms larger than 25 mm had a 6% rupture rate in the first year. In group 2, the cumulative rate of rupture was about 0.5% per year for lesions smaller than 10 mm and about 1% for those larger than 10 mm in diameter. In group 1, in addition to size, location was related to hemorrhage risk, with basilar tip, vertebral-basilar, posterior cerebral, and posterior communicating artery aneurysms having a higher risk for rupture. In group 2, only location (i.e., basilar tip) and increasing age predicted an increased rate of hemorrhage.

TABLE 360-2 Risk of Rupture of Unruptured Aneurysms

The strengths of the ISUIA include its multicenter design, which minimizes referral and treatment bias, and its size, which provides robust statistical power to formulate conclusions.¹⁴ The study has, however, been challenged on a number of points that are largely related to selection bias, the retrospective nature of the study, and the inclusion of patients with cavernous aneurysms in the study population.^1,¹⁵^–¹⁷ Cavernous carotid aneurysms are known to have a lower risk for hemorrhage.¹⁸ However, after excluding intracavernous aneurysms, the investigators concluded that the rate of hemorrhage was only slightly increased from 0.05% to 0.066% per year for small aneurysms and from 0.95% to 1.38% for large aneurysms.¹⁹

The most significant criticism of the study is related to the possibility of selection bias. Does the population studied truly represent a population of patients with UIA, or has some selection bias created a population with an inherently lower risk for rupture? This concern is particularly significant because selection bias cannot be corrected with any statistical methods. In regard to selection bias, all patients were selected for observation or surgery after consultation with a neurosurgeon.^1,¹⁵^–¹⁷ ^,20 If it is assumed that most experienced neurosurgeons have an intuitive concept of what constitutes an aneurysm at high risk for rupture (e.g., size, configuration, family history), it is reasonable to presume that high-risk patients were treated and removed from the study pool. Removal of these high-risk patients could potentially skew the risk for rupture downward. Indeed, a calculation by Winn and associates based on the low rupture rate in the ISUIA project resulted in an extraordinary high UIA prevalence rate of 16% to 33%.^1,²⁰ This calculated prevalence rate derived from ISUIA data is several magnitudes larger than the reported prevalence rates of UIAs² (see Table 360-1).

• Juvela and coworkers studied 142 Finnish patients with UIAs and a mean follow-up duration of 19.7 years.²¹ Of the original population, 131 (92%) had previously experienced SAH, were found to have multiple aneurysms, and had only the offending lesion clipped. The investigators found an annual incidence of about 1.3% (95% confidence interval [CI], 0.9% to 1.7%). Seventy percent of the aneurysms that ruptured were also smaller than 6 mm, and the aneurysm rupture rate was found to increase linearly with lesion size. The study by Juvela and coworkers²¹ is unique because the confined catchment area and stable Finnish population made a long and a high follow-up percentage possible.¹⁶

As with other studies, the analysis by Juvela and colleagues ²¹ has limitations. Although the study generally lacked selection bias, the aneurysm population was derived from 30 to 50 years ago, before the introduction of imaging techniques, and it may represent a different group of patients than seen today.²² In addition, the data were compiled from a single center with the inherent single-center bias.²³ A potential does exists for genetic bias because the Finnish population is known to have a higher prevalence of UIAs, a higher incidence of SAH (13 to 16 per 100,000 persons per year^24,²⁵) than in other Western countries (10/100,000 persons per year²⁶), and a different aneurysm distribution (higher frequency of middle cerebral artery aneurysms).^17,²⁷ Most importantly, with regard to the validity of the natural history of UIAs, the study had insufficient numbers of truly incidental asymptomatic aneurysms; only 5 (4%) of the 142 patients had no previous history of SAH.²¹ The patients in the study by Juvela and associates²¹ were similar to those reported by Winn and coauthors⁷ in that the study involved long-term follow-up of patients with multiple aneurysm who suffered an SAH and had the offending lesion clipped, as well as similar to the patients in group 2 of the ISUIA study. All three studies reported a rate of rupture approximating 1%.

• Tsutsumi and colleagues reported 62 patients with UIAs treated conservatively at locations not related to the cavernous sinus.²⁸ The mean follow-up was 4.3 years. Seven patients (11.3%) experienced SAH confirmed by CT. The 5- and 10-year cumulative risk for CT-confirmed SAH calculated by the Kaplan-Meier method was 7.5% and 22.1%, respectively, for total cases; 33.5% and 55.9%, respectively, for large (>10 mm) aneurysms; and 4.5% and 13.9%, respectively, for small (<10 mm) aneurysms.²⁸

However, this study by Tsutsumi and colleagues ²⁸ must be interpreted with caution because the sample population was small and the power of the statistical analysis is therefore in question. Moreover, the follow-up period was relatively short, and the data were derived from a single center, with the associated single-center bias. The patients were obtained from a population of patients undergoing cerebral angiography who were older (mean age, 70.8 years) and had a high concomitant rate of ischemic and hemorrhagic events.

In summary, the yearly rupture rate of UIAs remains to be defined precisely, but in general it is 1% for aneurysms approximately 10 mm in size.

Factors Associated with Rupture

As already noted, the rate of rupture of UIAs may be affected by multiple factors related to the aneurysm or the patient.

For example, in 1974 Mount and Brisman reviewed 158 patients with unruptured, multiple aneurysms (which included the earlier study of Heiskanen and Marttila ³⁰) monitored for an average of 5 years and noted a bleeding rate of at least 2% per year.³⁰ Wiebers and associates also found that multiple aneurysms had a greater propensity for rupture than did solitary aneurysms.^6,⁸ This was consistent with the data from Winn and colleagues.⁷ Yasui and coworkers demonstrated an annual rupture rate of 6.8% in patients with multiple aneurysms and 1.9% in those with single UIAs.¹³ A meta-analysis by Rinkel and associates found that the risk for rupture was higher in patients with multiple aneurysms, with a 1.7 relative risk for rupture in patients with multiple lesions in comparison to those with an asymptomatic lesion.² In contrast, a study conducted in Helsinki by Juvela and colleagues with 2 decades of follow-up did not confirm a higher risk for SAH in patients with multiple UIAs.^10,²¹

In summary, most studies support the concept that multiple UIAs are associated with a higher risk for hemorrhage than solitary aneurysms are.

Aneurysm Growth

The report by Juvela and coauthors of patients harboring a UIA had a median follow-up period of 14 years and noted that aneurysms that subsequently ruptured (17 patients) displayed a significant increase in size.¹⁰ Among the 14 patients for whom angiographic follow-up was available and in whom there was no sign of rupture, no significant increase in aneurysm size was noted. Interestingly, growth was strongly associated with cigarette smoking (odds ratio [OR], 3.48; 95% CI, 1.14 to 10.64; P < .05).¹⁰ In contrast, Sampei and coworkers observed the growth of aneurysms between successive angiographic examinations in 25 patients and noted that rebleeding did not appear to be affected by the growth rate or by the initial size of the aneurysm.³¹ However, 11 of the patients were monitored for only less than 1 month.

In summary, the data in the literature are insufficient to conclusively document a relationship between aneurysm growth and risk for rupture.

Symptomatic Aneurysms

Symptomatic aneurysms are aneurysms accompanied by signs and symptoms related to the lesion, excluding clinical features related to SAH. The symptoms may be mild, such as headaches, or more severe, such as cranial nerve palsies and brainstem signs. The studies of Locksley ³ and Rinkel and colleagues² support the existence of a relationship between symptomatic UIAs and an increased propensity for rupture. In the Cooperative Aneurysm Study, 34 patients with symptomatic, unruptured aneurysms were observed for almost 4 years (47 months), and 26% died of SAH (≈7%/yr).³ This rate was significantly higher than the rupture rate for incidental aneurysms (0.8%/yr). Rinkel and colleagues found that the relative risk for rupture of a symptomatic aneurysm was 8.2 times that of an asymptomatic lesion.² However, in a multivariate analysis by Wiebers and associates, no correlation could be found between risk for hemorrhage and symptoms.⁶ This was also the case for Juvela and coworkers, who found that the percentage of patients without SAH was 65% to 80% at 20 years of follow-up, which did not differ significantly when symptomatic and nonsymptomatic (i.e., incidental) aneurysms were compared.¹⁰

Excluding headaches as a symptom, the majority of symptomatic aneurysms are associated with cranial nerve III dysfunction and are therefore most likely located at the posterior communicating location. This location has been found to have a higher rate of rupture of UIAs.^14,³²^–³⁴ Moreover, to affect cranial nerve III function, a UIA must enlarge. Increased size has been demonstrated to be correlated with hemorrhage. Thus, the perception that symptomatic UIAs have an increased rate of rupture may be an epiphenomenon related to aneurysm location and size.

In summary, the data are equivocal on an association between symptoms and rupture in UIAs.

Only gold members can continue reading. Log In or Register to continue