The management of unruptured intracranial aneurysms (UIAs) has become more challenging for clinicians. In the past, all UIAs were considered for treatment to prevent the most devastating consequence of aneurysm rupture, subarachnoid hemorrhage (SAH). Because SAH is associated with an overall mortality of 50% and major morbidity of 30%,¹ significant treatment risks were often justified to prevent these poor outcomes. More recent data, however, suggest that risks of treatment are not without consequence, and the risk of aneurysm rupture may be lower than expected.

As our diagnostic imaging technology improves, the ability to detect incidental UIAs increases. To add to the complexity of designing rational treatment guidelines, individual patient characteristics as well as aneurysm characteristics may affect the natural history and alter the risks of intervention. The appropriate treatment for UIAs is also controversial, with endovascular coiling as an increasingly employed alternative to surgical clipping. Given all these issues, treatment decisions by a multidisciplinary team, comprising a vascular or intensive care neurologist, a vascular neurosurgeon, and a neurointerventionist is essential. Increasing evidence suggests that the availability and experience of surgeons and endovascular specialists factor into the risks and cost-effectiveness of treatment for these patients. Outcomes are also better at centers with higher volumes that provide the expertise of a multidisciplinary team.

The challenge facing clinicians today is balancing the risk of treatment designed to prevent SAH with the natural history risk of rupture. Clinicians need to be more careful in deciding which patient is appropriate for intervention. If local expertise is limited, transfer to an experienced center with a multidisciplinary team is recommended.

Epidemiology

Unruptured intracranial aneurysms are common, with a reported prevalence ranging from 0.2 to 8.9% of the general population. In a systematic review, retrospective autopsy studies reported the lowest rates and prospective angiography studies the highest rates, likely reflecting the different study designs and patient populations. For adults without specific risk factors, the estimated prevalence was 2.3%.² The majority of UIAs are asymptomatic and are incidentally discovered in evaluation for headaches and other neurological symptoms. Less than half of patients present with symptoms from compression of neighboring nerves or brain or from embolic events distal to the aneurysm site.^3,4 Detection of incidental UIAs has increased with improvement of noninvasive vascular imaging. Magnetic resonance angiography (MRA) and computed tomographic angiography (CTA) are used increasingly as alternatives to digital subtraction angiography (DSA) (the “gold standard”). However, they have difficulty detecting aneurysms less than 5 mm in size, which make up a third of all aneurysms.⁵ Sensitivity for CTA and MRA was 76 to 98% and specificity was 85 to 100%, in a population with SAH, but this may be an overestimation of the accuracy of these techniques.⁶ Given these limitations, screening with noninvasive measures for aneurysms is not cost-effective but may have a limited role in those with specific risk factors.⁷

Risk factors for the formation of intracranial aneurysms include two first-degree relatives with cerebral aneurysm, any first-degree relative with history of SAH, age greater than 50 years, female gender, smoking, cocaine use, and inherited disorders such as polycystic kidney disease.^2,8 Patient characteristics such as smoking, younger age, female gender, prior SAH, and hypertension have been associated with increased risk of aneurysm rupture. Active smoking status was associated with a relative risk (RR) of 1.46, increasing to RR 3.04 with time.⁹ Smoking and female gender were also associated with increased risk of aneurysm growth and de novo aneurysm formation.¹⁰

Based on prior studies, risk of rupture may also be influenced by aneurysm characteristics such as location, morphology, and size. In retrospective reviews, location in the anterior circulation, size greater than 10 mm, and multiple aneurysms were associated with higher rupture rates.^9,11–14 Aneurysm size has been reported as the main predictor of risk of subsequent rupture in most natural history studies. Risk factors for increasing aneurysm size included multiple aneurysms, female gender, age over 70 years, and anterior communicating and basilar artery location.¹²

Natural History

Until recently, the natural history of UIAs has been based primarily on case reports and case series. Rupture rates have been estimated at 1 to 2% per year, varying widely from studies. The highest annual rupture rates of 3.2% are from a Japanese national hospital registry of 427 patients with 217 person years of follow-up.¹⁵ The Helsinki study which reported 1.4% annual rupture rate benefited from longer follow-up times (2575 person years) without surgical selection but tended to represent a younger patient population in a single center.¹⁰ Size was the biggest predictor of rupture, along with multiple aneurysms in younger patients.⁹

Our best data on the natural history of UIAs comes from the prospective multicenter International Study of Unruptured Intracranial Aneurysms (ISUIA), which involved 61 centers enrolling patients from 1991 to 1998 with nearly 1700 patients and 6544 person years of follow-up.¹⁶ To date, this is the largest and most comprehensive natural history study on UIAs published. Both the retrospective and prospective arms of this study have generated much controversy because the reported rupture rates are much lower than expected, and the risks of treatment were not without consequence. In the retrospective arm of ISUIA, patients were divided into subgroups based on previous history of SAH. For aneurysms less than 10 mm, the annual rupture rate was 0.05% per year in Group I (no SAH) and 0.5% per year in Group II (with SAH).¹⁷ In the prospective study, the overall incidence of aneurysm rupture was 0.8% per year with a mean follow-up time of 3.9 years. Risk of rupture was highest with aneurysms located in the posterior circulation and posterior communicating artery (RR 2.3), similar to findings in the retrospective study. Size greater than 7 mm in diameter, history of prior SAH, and younger age was also associated with an increased risk of rupture.¹⁶

The main limitation of ISUIA was lack of randomization. Patients were selected for treatment based on clinicians’ preference. Treatment of selected patients introduces bias that could lead to underestimating the rupture rates. Symptomatic aneurysms tend to be underrepresented in most natural history studies, including ISUIA, because they are often preferentially treated. The most common symptoms were cranial neuropathies, ischemic changes from embolic events, and visual loss.⁴ Nonetheless, ISUIA represents our best natural history data and suggests that treatment may not be justified in patients with small, asymptomatic aneurysms in the anterior circulation without prior history of SAH.

Treatment

The goal of treating UIAs is to reduce the risk of rupture and the devastating consequences of SAH. As with all treatment interventions, the risks of treatment should be balanced against the risk of rupture if UIAs are left untreated. The American Heart Association (AHA) Stroke Council recommends treatment for all symptomatic intradural aneurysms regardless of size, aneurysms in patients with prior SAH, and asymptomatic aneurysms greater than 10 mm.¹⁸ In light of recent data, however, critical assessment of factors such as aneurysm size, location, and morphology; patient age and comorbidities; and experience of treating physicians must also be considered. Although many UIAs would be considered for treatment, conservative management may be a reasonable choice for some patients. For patients who do not undergo treatment, serial imaging along with risk factor modification with smoking cessation and control of hypertension is warranted. Serial imaging, typically with MRI or CT, within the first year after diagnosis can be reassuring to document stable aneurysm size, but it is likely not a reliable means to detect acute changes in aneurysm size associated with rupture.

Surgical Clipping

Once the decision to treat UIAs is made, the next dilemma is choosing the appropriate treatment modality, surgical clipping or endovascular coiling. Surgical clipping has been the mainstay of aneurysm treatment since the first successful clipping was performed by Walter Dandy in 1937. Microsurgical techniques have continued to evolve, but reported rates of surgical morbidity and mortality vary widely among case series. In the largest meta-analysis, consisting of 2460 cases with UIAs published in 61 reports, the mortality rate was 2.6% and the permanent morbidity rate was 10.9%.¹⁹ The most-comprehensive data came from the ISUIA study of 1971 clipped patients with an average 4 year follow-up including functional and cognitive outcomes. At 1 year follow-up, mortality was 3.2% and neurological disability was 12%, with a combined surgical morbidity and mortality of 15.2%.¹⁶ Risk factors for poor outcome in surgical patients included patient age greater than 50 years, aneurysm size greater than 12 mm, location in the posterior circulation, calcified or atherosclerotic aneurysms, and presence of symptoms.^16,19

Endovascular Coiling

Since the development of the Guglielmi detachable coil, coil embolization has been used with increasing frequency. Early cases were often selected for endovascular treatment because of poor surgical risk. Recent technological advances may not be reflected in prior studies of outcomes after coil embolization. In a recent metaanalysis of 90 UIAs treated with coil embolization, annual mortality was reported as 1.4% with total procedural morbidity and mortality of 8.1%.²⁰ Endovascular coiling of 1811 aneurysms, both ruptured and unruptured, at a single center reported early mortality of 1.5% and morbidity of 5.3%.²¹ The ISUIA study, which included 451 coiled patients with less than 4 years of follow-up, reported a mortality of 3.4% and total morbidity and mortality of 9.8% at 1 year, similar to surgically treated patients. Direct comparison of the two groups was not possible given lack of randomization and significant differences in patient age, aneurysm size, and location between treatment groups. Risk factors for endovascular coiling included larger aneurysms in the posterior circulation.¹⁶

Comparative Studies

Unfortunately, no randomized trials of treatment for UIAs have been performed to determine the indications for clipping versus coiling in this patient population. A small, randomized trial of 109 patients with ruptured aneurysms at a single center failed to show any significant differences in outcome after 3 months or 1 year.²² The only other direct comparison between clipping and coiling is from the International Subarachnoid Aneurysm Trial (ISAT). The trial found a significant 23% reduction of relative risk (7% absolute risk reduction) in dependency or death at 1 year with endovascular treatment.²³ These trials included patients with ruptured aneurysms; hence, the results cannot be directly extrapolated to the UIA population.

In a retrospective cohort study using the National Inpatient Sample Data from 1996–2000, short-term outcomes were examined for patients with UIAs undergoing clipping versus coiling. There were far more patients clipped during this period, and clipped patients were slightly younger. There were no differences in mortality and discharge to long-term care; however, coiled patients were discharged home more frequently, had a shorter length of stay, lower hospital costs, and less neurological complications. This study was limited in determining long-term efficacy and functional outcomes.²⁴ In a cohort study of 2612 patients with UIAs treated at 70 university hospitals between 1994 and 1997, hospital death, and discharge to nursing home or rehabilitation center were more common in surgical- compared with endovascular- treated patients (18.5% vs. 10.6%, p =0.002).²⁵ Similarly, a study of 2069 patients with UIAs in California treated between 1990 and 1998 reported significantly less adverse outcomes with endovascular compared with surgical treatment (10% vs. 25%, p <0.001).²⁶ In a singlecenter cohort study, similar findings were reported including a longer follow-up of 3.9 years, suggesting persistent new symptoms or disability in surgical cases, 34% versus 8%, and a longer recovery period, 1 year versus 27 days for 50% return to normal.²⁷ These observational studies all consistently suggest that endovascular coiling is safer than surgical clipping for UIAs but are limited by the possibility that patient selection may be contributing to outcome differences.

In a recent prospective observational study of embolization and surgical clipping, surgical clipping was associated with short-term functional deficits that improved by 1 year, while coiled patients had no deficits in these areas. When outcome measures such as quality of life and mood are assessed, coil embolization also had more favorable outcomes, with surgery having a considerable but often reversible effect in the first 3 months postprocedure.²⁸ Coiled patients had significantly better short-term outcomes compared with clipped patients, particularly older patients greater than 65 years of age.²⁴

Efficacy of Treatment

The primary goal of treatment of UIAs is to prevent aneurysm rupture and the devastating consequences of SAH. Rupture rates after aneurysm treatment may give us the best estimate of treatment efficacy. Unfortunately, many studies fail to report the duration of follow-up needed to calculate these rates. In a recent case series including both ruptured and unruptured aneurysms, the risk of hemorrhage after surgical clipping ranged from 0.2 to 1.5% per year.^29–31 Incomplete aneurysm occlusion was associated with increased risk of rupture.³¹

The main criticism of endovascular coiling has been the durability of treatment. In a Scientific Statement from the AHA, six studies of coil embolization of ruptured aneurysms had a combined rerupture rate of 0.9% per year. Coil embolization had generally higher rupture rates, ranging from 0.6 to 3.0% per year, when compared with surgical clipping.³² Larger aneurysms and incompletely occluded aneurysms had more frequent aneurysm growth and increased risk of rupture after treatment.³³ A significant number of patients do not achieve complete occlusion after the first endovascular treatment, up to 50 to 78% in reported studies.^6,16,20 Long-term follow-up for durability of coiled patients will be important. Based on this limited evidence, both treatment modalities appear to reduce the risk of rebleeding, but surgical clipping may be more effective.

Treatment Experience

Higher-volume hospitals had fewer adverse outcomes than hospitals that treat a limited number of UIAs. Similarly, high-volume surgeons had significantly lower morbidity and modestly lower mortality rates.³⁴ The higher-volume centers had shorter lengths of stay and lower total hospital charges, suggesting better cost effectiveness of treatment.³⁵ In addition, the propensity for a hospital to use endovascular therapy was also associated with better outcomes. These studies suggest that selective referral of UIAs to centers with higher volume and with experienced neurosurgeons and interventional neuroradiologists could improve outcomes.³⁶

Conclusion

The natural history of UIAs is better defined with the results of large cohort studies like ISUIA. The current literature exposes the complexity of decision-making for clinicians when presented with asymptomatic patients with UIAs. Treatment decisions, including medical observation, require balancing the risk of intervention with the natural history of risk of rupture for each individual patient. Clinicians must rely on judgment and expertise given the scarcity of evidence to help guide these decisions.

The choice of treatment modality, surgical clipping or endovascular coiling, should be individualized. Specific characteristics of the patient and the aneurysm influence the risks of treatment and the likelihood of complete obliteration of the aneurysm. Aneurysm location, size, and morphology often determine the best therapy. Selecting the best treatment modality requires incorporating information about the patient’s age and medical condition, the aneurysm characteristics, and the technical ability of the practitioners. Given the absence of randomized trials in the treatment of UIAs, no specific recommendations can be made about the indications for surgical clipping and endovascular coiling. Direct comparison of treatments in the observational studies is limited by lack of randomization with differences in patient characteristics. A randomized trial is needed to define the best treatment for UIAs. The majority of evidence suggests that endovascular coiling is safer than surgical clipping; it should be considered preferentially in older patients with posterior circulation aneurysms. Surgical clipping may be a more durable treatment and be preferred by patients who desire a single treatment procedure, have wide-necked aneurysms, and have aneurysms arising in the middle cerebral artery.

Advances in imaging techniques that can evaluate aneurysm flow and hemodynamics, improvements in coiling with stents, microsurgical techniques, and combined surgical and endovascular treatments will likely improve outcomes for patients with UIAs. A multidisciplinary approach, with input from practitioners with the appropriate expertise, is probably the most reliable way to make the best treatment decisions. Patient outcomes are better at hospitals that offer both surgical clipping and endovascular therapy regardless of whether a patient is ultimately treated with surgery or coil embolization.