De Novo Aneurysm Formation and Growth of Untreated Aneurysms
A 5-Year MRA Follow-Up in a Large Cohort of Patients With Coiled Aneurysms and Review of the Literature
Background and Purpose—Rates of development of de novo intracranial aneurysms and of growth of untreated additional aneurysms are largely unknown. We performed MRA in a large patient cohort with coiled aneurysms at 5-year follow-up.
Methods—In 276 patients with coiled intracranial aneurysms and 5±0.5 years of follow-up MRA (totaling 1332 follow-up patient-years), additional aneurysms were classified as unchanged, grown, de novo, or incomparable with previous imaging. We calculated 5-year cumulative incidence of de novo aneurysm formation and growth of untreated aneurysms. We searched PubMed and EMBASE databases for studies assessing aneurysm development, and growth.
Results—In 50 of 276 patients (18%), 75 additional aneurysms were present at follow-up MRA. Of these 75, 2 were de novo (both 3 mm), 58 were unchanged, 5 had grown from 1 to 3 mm (7.9% of 63 known additional aneurysms; 95% CI, 1.3%–14.6%), and 10 were incomparable. Five-year cumulative incidence for a de novo aneurysm developing was 0.75%. Four additional aneurysms in 3 patients were treated. Ten previous studies reported annual incidences of growth of additional aneurysms ranging from 1.51% to 22.7%, and 5 studies reported annual incidences of de novo aneurysm formation ranging from 0.3 to 1.8%.
Conclusions—MRA screening of patients with coiled aneurysms within the first 5 years after treatment has a low rate of de novo aneurysm development and growth of additional aneurysms, and an even lower treatment rate.
In patients with a symptomatic intracranial aneurysm, additional aneurysms are frequently found and new aneurysms may develop with time.1,–,5 Additionally found aneurysms that are small are often left untreated because of unfavorable location or geometry, together with a low anticipated chance of rupture.
The incidence of development of de novo aneurysms and the rate of growth of untreated additional aneurysms have been the subject of several studies. These studies differ in design and results. An important drawback is the lack of fixed follow-up intervals in most studies that impedes assessment of incidence of development and growth of aneurysms. Because growth rate of aneurysms in not constant over time,6 recalculating annual incidences from the total period of follow-up is inaccurate.
Proven aneurysm growth is considered a risk factor for rupture because aneurysm size is directly related to the risk of rupture, and growing aneurysms may be regarded as unstable.7,8 We assessed the incidence of de novo aneurysm formation and determined the natural history of additional untreated aneurysms in a large cohort of patients with coiled intracranial aneurysms after a follow-up period of 5 years. In addition, we provide an overview of other relevant studies in relation to our results.
Patients and Methods
Institutional Review Boards of the 7 participating medical centers (Academic Medical Center Amsterdam, Leiden University Medical Center, Maastricht University Medical Center, Slotervaart Ziekenhuis Amsterdam, St. Elisabeth Ziekenhuis Tilburg, University Medical Center Utrecht, and VU Medical Center Amsterdam, all in the Netherlands) approved the study protocol. All participants provided written informed consent.
From the databases of the centers, we retrieved all patients with a ruptured or unruptured intracranial aneurysm that was coiled since January 1995. Patients with adequate aneurysm occlusion (complete occlusion or a small neck remnant) at 6-month angiographic follow-up, according to occlusion status recorded in the databases and radiological reports, were selected to participate in a study to assess long-term stability of coiled intracranial aneurysms (LOTUS). Inclusion criteria for LOTUS were follow-up duration >4.5 years, current age between 18 and 70 years, independent functional state, and no contraindications for 3-T MRI.
Eligible patients received a letter with an invitation to participate in LOTUS. Patients who did not respond to the invitation letter were contacted by telephone.
For the purpose of the present study, we selected those patients included in the LOTUS study who had a follow-up MRA 5±0.5 years after treatment. Our group has previously reported the first 65 patients.9 We extended the number of patients to obtain more reliable data.
MR Imaging Follow-Up Protocol
MR imaging examinations were performed centrally in 2 centers on a 3-T system (Intera R10; Philips Medical Systems) by using the sensitivity encoding (SENSE) phased-array head coil (MR Imaging Devices). MR imaging protocol included axial T2-weighted fast spin-echo and multiple overlapping thin-slab acquisition 3-dimensional time-of-flight MRA sequences. Detailed descriptions of the imaging parameters have been described previously.9 Images were processed into maximum intensity projections and volume-rendered 3-dimensional images of the circle of Willis. Total MR imaging examination time was 20 minutes.
MR Imaging and MRA Evaluation
MR images were evaluated independently by 2 experienced neuroradiologists in 3 of the participating centers, with discrepancies resolved in consensus. Presence, location, and largest diameter of additional aneurysms were compared with conventional CT or MRA at the time of coiling or at 6-month follow-up. Additional aneurysms were classified as unchanged, grown, de novo, or incomparable with previous imaging.
Proportions of patients with additional aneurysms, patients in whom a de novo aneurysm developed, and proportions of previously known additional aneurysms that had grown were calculated with 95% CI. Characteristics of patients with additional aneurysms were compared with patients without additional aneurysms. Patient and aneurysm characteristics of patients with grown additional aneurysm were compared to those of patients with no aneurysm growth. The sample t test was used for comparison of means and relative risks were calculated for proportions. Clinical implication of the 5-year-follow-up MRA in terms of treatment advice and change of follow-up imaging policy were described.
Literature Search and Study Eligibility
We searched PubMed and EMBASE databases until April 2010. The following key words as MESH terms and text words were used: “intracranial aneurysm” in combination with “AND” “additional,” “growth,” “new,” and “novo,” and in “OR” combinations. Studies were included if prevalence or incidence of growth of untreated intracranial aneurysms or de novo aneurysm formation was assessed and if sample size was >50. To assess eligibility, the reviewer (S.P.F.) screened titles and abstracts and reviewed relevant full-text articles on inclusion criteria. Reference lists of relevant studies were searched for additional studies.
Patient and Aneurysm Characteristics
Of 1808 intracranial aneurysms in 1675 patients with coiling in the 7 participating centers in the Netherlands, 1287 with 1412 aneurysms had 6-month follow-up angiography and 1066 (75%) aneurysms in 971 patients were adequately occluded at this first angiographic follow-up. Of these 971 eligible patients with 1066 aneurysms, 157 (16%) could not be traced, and 274 could not be included for a variety of reasons (Figure 1). The remaining 540 patients were invited to participate in the study and 140 declined (participation grade, 74%).
Of 400 patients with 440 coiled aneurysms participating in LOTUS, 75 patients (19%) had 101 additional aneurysms. Of 400 patients, 276 with 300 coiled aneurysms had a follow-up duration of 5 years±6 months (totaling 1332 follow-up patient-years); the follow-up interval of the remaining 124 participants was 5.5 to 12.9 years. The 276 patients with 300 coiled aneurysms and 1332 follow-up patient-years are the subject of this study. Fifty of these patients had 75 additional aneurysms. Patient and aneurysm characteristics of all 276 patients and the 50 patients with additional aneurysms are displayed in Table 1.
Additional Aneurysms on 5-Year Follow-Up MRA
In 50 of the 276 patients (18.1%; 95% CI, 13.6%–22.7%), 75 additional aneurysms were found. Additional aneurysms were more often present in women (relative risk, 2.03; 95% CI, 1.03%–3.98); other patient characteristics were comparable. Thirty-six patients had 1 additional aneurysm, 8 patients had 2 additional aneurysms, and 6 patients had ≥3 additional aneurysms. Characteristics of 75 additional aneurysms are displayed in Table 2.
Ten of 75 additional aneurysms (13%) were classified as incomparable. No previous imaging was available in 7 additional aneurysms, and in another 3 the projection of the initial angiogram did not allow verification of their presence. The remaining 65 additional aneurysms could be compared with previous imaging; 58 (89.2%; 95% CI, 81.7%–96.8%) were unchanged and 2 (3.1%; 95% CI, 0.8%–10.6%) were de novo. Five of 63 previously known untreated aneurysms had grown (7.9%; 95% CI, 1.3%–14.6%). The 58 unchanged additional aneurysms were present in 41 patients. Sizes ranged from 1 to 7 mm.
The 5 additional aneurysms in 4 patients that had grown had the following characteristics. The first patient, a 45-year-old man, had 5 additional aneurysms, of which a pericallosal artery aneurysm had increased in size from 1.5 to 2.5 mm.9 The second patient, a 49-year-old woman, also had 5 additional aneurysms, of which a right internal carotid artery tip aneurysm had increased from 3 to 6 mm (Figure 2). This aneurysm is scheduled for coiling. The third patient, a 66-year-old man, had a posterior communicating artery aneurysm that increased in size from 2 to 3 mm. The fourth patient, a 53-year-old woman, had 2 additional aneurysms that both had increased in size: a posterior inferior cerebellar artery aneurysm had grown from 2 to 3 mm and a basilar tip aneurysm had grown from 4 to 6 mm. The basilar tip aneurysm was subsequently coiled. Patient gender and age were not predictors of aneurysm growth (Table 2). Size of the additional aneurysm also was not a predictor of aneurysm growth; mean size of aneurysms without growth was 2.6 mm and mean sized of aneurysms with growth was 2.8 mm.
Five years after coiling, 2 de novo aneurysms were found in 2 of 276 patients (0.7%; 95% CI, 0.2%–2.6%): 1 3-mm middle cerebral artery aneurysm in a 56-year-old woman and 1 3-mm basilar tip aneurysm in a 66-year-old man. Cumulative 5-year incidence of de novo aneurysm formation was 0.75% (95% CI, 0.2%–2.7%), ie, 2 de novo aneurysms in 1332 patient-years.
Clinical Implications of 5-Year MRA Follow-Up
Of 75 additional aneurysms, 2 left middle cerebral artery aneurysms in 1 patient classified as incomparable with previous imaging were clipped. Two grown additional aneurysms were coiled in 2 patients: 1 on the internal carotid artery tip and 1 on the basilar tip. Thus, after 5-year MRA follow-up, 3 of 276 patients were advised treatment for an additional aneurysm (1.1%; 95% CI, 0.4–3.1). Both de novo aneurysms were left untreated and follow-up imaging is scheduled. In 8 patients with 8 additional aneurysms (2.9%; 95% CI, 0.9%–4.9%), treatment was judged not indicated by the treating multidisciplinary team because of small size, but extended follow-up imaging was advised. Other patients with additional aneurysms were not planned for extended imaging follow-up, because it was judged unnecessary by the treating multidisciplinary team in the clinical context of the individual patient.
Other Studies Reporting De Novo Aneurysms and Growth of Untreated Aneurysms
Searching the literature yielded 12 relevant studies (Table 3).1,–,5,9,–,15 Three studies of 610, 102, and 112 patients with clipped aneurysms reported additional aneurysms on imaging follow-up,1,3,4 1 study reported additional aneurysms on imaging follow-up of 65 patients with coiled aneurysms,9 and 8 studies reported additional aneurysms on imaging follow-up in selected cohorts of 57 to 321 patients with known untreated additional aneurysms.2,5,10,–,15
Proportions of growth of additional aneurysms were described in 10 studies with backward calculations of these proportions to annual risks, which ranged from 1.51% to 22.7%. Growth rate of additional aneurysms was assessed in 2 studies, with a recalculated annual rate of 0.31 and a range of 0.12 to 1.3 mm. Risk factors for aneurysm growth were assessed in 9 studies. Additional aneurysm sizes ≥5 mm,10,13 ≥8 mm,12 and ≥10 mm14 were predictors for growth in 4 studies. Patients with multiple additional aneurysms had higher chance of aneurysm growth in 3 studies,1,11,13 smoking was a predictor for aneurysm growth in 2 studies,1,2 and female gender was a predictor for aneurysm growth in another 2 studies.2,11 The following predictors for aneurysm growth were found in 1 study each: advanced patient age,11 patient5 or family13 history of subarachnoid hemorrhage, alcohol abuse,15 and aneurysms with multiple lobes.10 Each of the 4 large vessels have been suggested to be at higher risk for growth in different studies: the anterior cerebral artery,11 internal carotid artery,12 middle cerebral artery,13 and basilar artery.11,12
De novo aneurysm formation was reported in 5 studies, with recalculated annual risks ranging from 0.3% to 1.8%. Risk factors for new aneurysm formation were assessed in 4 studies. Smoking was a predictor for de novo aneurysm formation in 2 studies,1,2 and female gender,2 multiple aneurysms,3 and follow-up duration >9 years4 were risk factors in 1 study each.
Imaging follow-up 5 years after coiling to detect de novo aneurysm formation has a very low yield. The cumulative 5-year incidence in our cohort was only 0.75%. Some previous studies found higher annual incidences up to 1.8%.1,–,4 However, unlike our study, these studies had no fixed follow-up period but wide follow-up intervals, making it difficult to assess the true timing of development of de novo aneurysms. Three studies had a much longer follow-up of 9 to 19 years.1,2,4 As can be expected with longer follow-up, more de novo aneurysms were detected in these studies.
Growth of untreated aneurysms in our study was more frequent than development of de novo aneurysms, with a 5-year incidence of 7.9%. Other studies found widely ranging annual incidences of 1.51% to 22.7%. We could not find a risk factor for aneurysm growth, but in previous studies consistent risk factors for growth of additional aneurysms were aneurysm size (≥5, ≥8, and ≥10 mm in different studies) and presence of multiple aneurysms. In our series, all but 1 additional aneurysm were <7 mm. Two of 4 patients in our series with grown aneurysms had 5 additional aneurysms.
Two of 5 preexistent additional aneurysms that showed growth on 5-year follow-up MRA were subsequently coiled. Indication for treatment was decided in multidisciplinary meetings. Three grown aneurysms were not treated because of small size (all 3 mm), consistent with international guidelines based on the ISUIA study.8 If aneurysm growth in itself would be an indication for treatment, then treatment rate in our study would have been 2-times higher (2.2%).
The concern of growth of untreated aneurysms is the increased risk of rupture, because size is an important determinant of the risk of rupture, and possibly also because enlarging aneurysms are unstable.7,8 In a meta-analysis for the risk of rupture of unruptured intracranial aneurysms, annual incidence of subarachnoid hemorrhage from an unruptured aneurysm was 1.2% in studies with a mean follow-up duration of <5 years, 0.6% in studies with mean follow-up duration of 5 to 10 years, and 1.3% in studies with a mean follow-up duration of >10 years.16 Juvela et al17 more recently found a comparable annual rupture rate of 1.3%. Long-term follow-up data from the ISAT study suggest an annual rupture rate from de novo and known additional aneurysms in patients with a coiled aneurysm to be 0.036% each (3 events in 8447 person-years).18 Miller et al19 estimated the incidence of rupture of de novo aneurysms to be 0.06% per patient-year in patients with a clipped aneurysm.
Combining the data from our study and the low rupture risk from additional aneurysms suggests that the risk of de novo aneurysm formation and significant enlargement of additional untreated aneurysms is low, with a subsequent extremely low risk of subarachnoid hemorrhage from these aneurysms. This low risk seems particularly true for the first 5 years, and probably also for the first 10 years. Therefore, screening of all patients within the first 5 years after aneurysm treatment does not seem beneficial in terms of preventing subarachnoid hemorrhage or for detection of aneurysms that need treatment.
Cost-effectiveness analyses for periodic screening for detection of grown and de novo aneurysms in patients with a history of subarachnoid hemorrhage and clipping proved that screening was not cost-effective, mostly because of screening-induced fear in patients.20 A limitation of our study is that 13% of additional aneurysms (10 aneurysms in 7 patients) could not be compared with previous imaging and were considered not comparable. It is therefore possible that we underestimated the incidence of de novo aneurysm formation and growth of untreated aneurysms in our patient cohort. This problem will most likely not occur in future studies, because in current practice patients are more frequently followed-up with MRA, resulting in complete imaging of cerebral vessels. A second limitation is that the low rate of grown and de novo additional aneurysms did not allow for reliable identification of risk factors for aneurysm growth and formation of de novo aneurysms. Risk factors found in other studies are female gender, smoking, aneurysm size >5 mm, and the presence of multiple additional aneurysms.1,–,3,5,10,–,14
MRA screening of patients with coiled aneurysms within the first 5 years after treatment has a low rate of de novo aneurysm development and growth of additional aneurysms and an even lower treatment rate.
Sources of Funding
This work was supported by a grant from the Nuts Ohra Foundation (SNO0602-22), the Netherlands, and the Netherlands Brain Foundation (15F07.12).
- Received May 29, 2010.
- Accepted September 24, 2010.
- © 2011 American Heart Association, Inc.
- Wermer MJ,
- van der Schaaf IC,
- Velthuis BK,
- Algra A,
- Buskens E,
- Rinkel GJ
- Juvela S,
- Poussa K,
- Porras M
- Tsutsumi K,
- Ueki K,
- Morita A,
- Usui M,
- Kirino T
- Wermer MJ,
- van der Schaaf IC,
- Velthuis BK,
- Majoie CB,
- Albrecht KW,
- Rinkel GJ
- Wiebers DO,
- Whisnant JP,
- Huston J III,
- Meissner I,
- Brown RD Jr.,
- Piepgras DG,
- Forbes GS,
- Thielen K,
- Nichols D,
- O'Fallon WM,
- Peacock J,
- Jaeger L,
- Kassell NF,
- Kongable-Beckman GL,
- Torner JC
- Sprengers ME,
- van Rooij WJ,
- Sluzewski M,
- Rinkel GJ,
- Velthuis BK,
- de Kort GA,
- Majoie CB
- Burns JD,
- Huston J III,
- Layton KF,
- Piepgras DG,
- Brown RD Jr.
- Wermer MJ,
- van der Schaaf IC,
- Algra A,
- Rinkel GJ
- Molyneux AJ,
- Kerr RS,
- Birks J,
- Ramzi N,
- Yarnold J,
- Sneade M,
- Rischmiller J
- Wermer MJH,
- Koffijberg H,
- van der Schaaf IC