Subarachnoidhemorrhage (SAH) is a serious disease even today,1–4even though the rate of treatment of unruptured intracranial aneurysms (UIAs) is on the rise, with the aim at reducing its incidence.4–6The effect of treating UIAs on the incidence of SAH has been clearly lower7than that of reduced prevalence of smoking.7,8The majority of diagnosed UIAs never rupture during the patient’s lifetime, even in the case of working-age patients.9,10
The indications for treating UIAs are becoming more difficult to define when the incidence of SAH decreases and the discovery of UIAs increases. The best-known form of UIA rupture scoring is that based on a large meta-analysis of individual data for 8382 patients gathered from 6 prospective cohort studies.11This population, hypertension, age, size of aneurysm, earlier aneurysm rupture, site of aneurysm (PHASES) score showed those factors to predict an increased risk of aneurysm rupture during a mean follow-up of 3.5 years. Most patients in these cohorts were those left for conservative follow-up because of an estimated low rupture risk or a limited life expectancy because of advanced age, severe diseases, or high treatment risks.11It is not possible to study the true natural history of UIAs in future prospective follow-up cohorts, particularly among patients of working age, because an increasing proportion of their UIAs are already treated at baseline.4,11Furthermore, the validity of the PHASES score for predicting the lifelong UIA rupture risk is unknown.
A treatment score devised recently for the present cohort suggested that the lifelong risk of aneurysm rupture can be predicted with only a few variables known at baseline: age < 40 years, current smoking, and UIA location and size.12This cohort represents an almost lifelong prospective follow-up study of UIAs among patients of working age without treatment selection bias,1,8–10,13–15and it was considered to be of high quality among all the known UIA series.11,13The aim was to study the validity of the PHASES score in predicting the long-term UIA rupture risk compared with this new treatment score.12
开云体育世界杯赔率
Patient Population
This series comprised 142 White European patients (median age 42 years, 76 women) with 182 UIAs diagnosed before 1979 at the Department of Neurosurgery, Helsinki University Central Hospital (i.e., at the time when UIAs were not operated on in Finland). This hospital was responsible for neurosurgical services for nearly the whole Finnish population. For details of the cohort with the inclusion and exclusion criteria as well as follow-up arrangements, see the previous reports.1,9,10,12,14–17All patients gave their written informed consent. Approval for the surveys and the collection of follow-up data had been obtained from the local ethics committee in compliance with national legislation and the Declaration of Helsinki. The data that support the findings of this study are available from the corresponding author upon reasonable request.
Follow-Up Protocol
Detailed follow-up procedures have been reported previously.1,9,10,12,14–21Briefly, the follow-up was based on visits to outpatient clinics, telephone interviews, and postal questionnaires obtained from patients and family members approximately every 10 years since the early 1960s. The structured questionnaire included patient characteristics, health behavior, previous diseases and medication, and hospital visits. The last follow-up took place in 2012, when 20 patients were still alive without rupture of the UIA.9–12,17,19,21
Additional information was obtained from medical records of all the patients from other hospitals and general practitioners to double-check the accuracy of medical history and blood pressure (BP) levels. Autopsy reports and official death certificates were examined for all deceased patients. In Finland, a statutory medicolegal autopsy is performed on all those who die due to trauma or unknown causes (Act on Inquests into the Cause of Death, 459/1973, Finnish law). The follow-up was complete.
Risk Factors
Recently published recommended variables for the UIA studies were available for the present investigation,22and the PHASES score11and recently reported new treatment score (Table 1)12based on the present cohort were both recorded at baseline for each patient (Table 2）.A systolic BP repeatedly > 140 mm Hg, a diastolic BP > 90 mm Hg, or use of antihypertensive medication was used for the definition of hypertension. Current smoking was ascertained at baseline.
Predictors of Juvela score for aneurysm rupture risk
| Juvela Score* | Points |
|---|---|
| Age | |
| <40 yrs | 2 |
| ≥40 yrs | 0 |
| Cigarette smoking at baseline | |
| Yes | 2 |
| No | 0 |
| Site of aneurysm | |
| ACoA | 5 |
| ICA bifurcation | 4 |
| PCoA | 2 |
| Others | 0 |
| Size of aneurysm | |
| ≥7.0 mm | 3 |
| <7 mm | 0 |
| Total sum of points | 0–12 |
This presentation of the Juvela score is adapted from Juvela S, Poussa K, Lehto H, Porras M. Natural history of unruptured intracranial aneurysms: a long-term follow-up study.Stroke. 2013;44(9):2414–2421,9and it is based on a long-term follow-up of patients diagnosed between 1956 and 1978.1,9,10,12,14–21
PHASES and new risk scores stratified by aneurysm rupture status
| Factor | PHASES Score Points | Present Score Points | Patients w/ SAH, n = 34 (24%) | Patients w/o SAH, n = 108 (76%) | All Patients, N = 142 (100%) |
|---|---|---|---|---|---|
| P: population (%) | |||||
| North American, European—except Finnish | 0 | 0 | 0 | 0 | 0 |
| Japanese | 3 | 0 | 0 | 0 | 0 |
| Finnish | 5 | 0 | 34 (24) | 108 (76) | 142 (100) |
| H: hypertension (%) | |||||
| No | 0 | 0 | 20 (22) | 71 (78) | 91 (100) |
| Yes | 1 | 0 | 14 (27) | 37 (73) | 51 (100) |
| A: age (%) | |||||
| <40 yrs | 0 | 2 | 20 (34)* | 38 (66) | 58 (100) |
| 40–60 yrs | 0 | 0 | 14 (17) | 70 (83) | 84 (100) |
| 61–70 yrs | 0 | 0 | 0 | 0 | 0 |
| ≥71–80 yrs | 1 | 0 | 0 | 0 | 0 |
| S: size of aneurysm (%) | |||||
| <7.0 mm | 0 | 0 | 24 (21) | 92 (79) | 116 (100) |
| 7.0–9.9 mm | 3 | 3 | 6 (38) | 10 (63) | 16 (100) |
| 10.0–19.9 mm | 6 | 3 | 3 (43) | 4 (57) | 7 (100) |
| ≥20 mm | 10 | 3 | 1 (33) | 2 (67) | 3 (100) |
| 艾凡:SAH早些时候从另一个动脉瘤(%) | |||||
| No | 0 | 0 | 3 (27) | 8 (73) | 11 (100) |
| Yes | 1 | 0 | 31 (24) | 100 (76) | 131 (100) |
| S: site of aneurysm (%) | |||||
| ICA | 0 | 0 | 1 (5) | 18 (95) | 19 (100) |
| PCoA | 4 | 2 | 10 (30) | 23 (70) | 33 (100) |
| ICA bifurcation | 0 | 4 | 3 (38) | 5 (63) | 8 (100) |
| MCA | 2 | 0 | 15 (23) | 49 (77) | 64 (100) |
| ACoA | 4 | 5 | 4 (50) | 4 (50) | 8 (100) |
| ACA + PerA | 4 | 0 | 1 (17) | 5 (83) | 6 (100) |
| Posterior circulation | 4 | 0 | 0 | 4 (100) | 4 (100) |
| Current cigarette smoking† | |||||
| No | NA | 0 | 8 (15) | 45 (85) | 53 (100) |
| Yes | NA | 2 | 20 (29) | 50 (71) | 70 (100) |
| Mean total score by aneurysm rupture (SD) | |||||
| No SAH | 4.19 (2.24) | 3.04 (2.23) | |||
| SAH | 5.32 (2.32)* | 5.32 (2.36)‡ | |||
| Median score (IQR)§ | |||||
| No SAH | 4 (3–5) | 2 (2–4) | |||
| SAH | 5.5 (3.3–6.8)* | 5 (4–7)‡ |
ACA = anterior cerebral artery; MCA = middle cerebral artery; NA = not applicable; PerA = pericallosal artery.
Population points (5 points) are subtracted from total PHASES scores, which has no effect on significance levels.
p < 0.05.
Current smoking value at baseline is missing for 6 of the 34 patients with an aneurysm rupture (18%) and for 13 of the 108 patients without a rupture (12%).
p < 0.001.
The IQRs represent the range between the 25th and 75th percentiles.
All angiography studies performed at baseline were reexamined by an experienced neuroradiologist who had no knowledge of the patients’ case histories.9,15–21The locations and maximum diameters of the UIAs were measured from standard projections of conventional angiograms at baseline.
Statistical Analysis
The data were analyzed with IBM SPSS Statistics for Windows version 25.0 (IBM Corp.). The PHASES score and the new treatment score variables for the present cohort at baseline were compared according to the patients’ future aneurysm rupture status by using the Pearson chi-square test or Fisher’s exact test. Sum scores, which were expressed as the mean ± SD and the median with interquartile range (IQR), were compared using the Mann-Whitney U-test and t-tests. Each patient was followed up until SAH, death from a reason other than SAH, occlusion treatment of the UIA (3 cases with a follow-up lasting > 24.4 years), or the last follow-up contact occurred. Cox proportional hazards regression analysis was used to test these scores as predictors of future rupture of diagnosed UIA. Wald statistics were used to estimate hazard ratios (HRs) and 95% CIs. The proportionality assumption was confirmed. The cumulative probabilities of aneurysm rupture were calculated according to the score groups by performing the Kaplan-Meier product-limit analysis.
The areas under the curve (AUCs) for the receiver operating characteristic (ROC) curves were calculated for both the PHASES and the Juvela prognostic treatment scores.12The C-statistic with 95% CIs was used to test the association of the total scores with the observed rupture rates, with the AUC between the scores being analyzed using the technique of Hanley and McNeil.23This statistical analysis was performed using MedCalc for Windows version 19.1.3 (MedCalc Software). A 2-tailed p value < 0.05 was considered statistically significant.
Results
Patient Characteristics and Follow-Up
During a total follow-up of 3064 person-years (range 0.8–52.3 years per patient), 34 of the 142 patients (24%) had an aneurysm rupture.12SAH的累积率为10 11%years and 30% at 30 years. The median follow-up time between diagnosis and a subsequent aneurysm rupture was 10.6 years (IQR 7.3–17.3 years), and the median follow-up for patients without a rupture was 24.4 years (IQR 14.6–35.1 years).12
PHASES and New Treatment Scores for the Patients
The PHASES and present score factors for the patients are shown according to their aneurysm rupture status inTable 2. The mean PHASES score was higher in those with a rupture than in those without (10.32 ± 2.32 vs 9.19 ± 2.24, p = 0.012). The difference was the same when population points (5 points) were subtracted from the PHASES score (5.32 ± 2.32 vs 4.19 ± 2.24, p = 0.012). The new score was also higher in those with an aneurysm rupture (5.32 ± 2.36 vs 3.04 ± 2.23, p < 0.001).
The AUC for the ROC of the present scoring for rupture was fair (0.755, standard error [SE] 0.050, 95% CI 0.657–0.853, p < 0.001), and the optimal cutoff point was obtained at ≥ 5 versus < 5 points for sensitivity (0.607), specificity (0.789), positive predictive value (0.459), and negative predictive value (0.872) (Table 3,Fig. 1）.The AUC for the ROC of the PHASES score was 0.674 (SE 0.059, 95% CI 0.558–0.790, p = 0.005), and the optimal cutoff point was obtained at ≥ 6 versus < 6 points for sensitivity (0.500) and specificity (0.811) when nonsignificant population points were subtracted from the score (Fig. 1）.auc的分数的区别not significant (p = 0.21). The AUC of PHASES including also patients with missing values for smoking was 0.641 (SE 0.054, 95% CI 0.535–0.747, p = 0.013), and the optimal cutoff point was obtained also at ≥ 6 versus < 6 points for sensitivity (0.441), specificity (0.778), positive predictive value (0.385), and negative predictive value (0.816) (Table 3）.Because all patients were Finns, omission of population points from the PHASES score (5 points) had no effect on the significance levels.
Risk sum score and cumulative incidence of aneurysm rupture
| Aneurysm Rupture Risk Score Points | No. of Patients | % Annual Rupture Rate* | % Cumulative Rupture Rates (95% CI) | |
|---|---|---|---|---|
| 10 Yrs | 30 Yrs | |||
| PHASES score† | ||||
| <2, <7 | 15 | 0.2 (1/450) | 0 | 7 (0–21) |
| 2–5, 7–10 | 88 | 0.9 (18/1936) | 7 (2–13) | 26 (16–37) |
| 6–7, 11–12 | 26 | 1.9 (9/464) | 16 (2–31) | 47 (22–72) |
| >7, >12 | 13 | 2.8 (6/214) | 33 (6–59) | 60 (26–93) |
| Total | 142 | 1.1 (34/3064) | 11 (5–16) | 30 (21–39) |
| Present score‡ | ||||
| 0 | 18 | 0 (0/513) | 0 | 0 |
| 1–4 | 68 | 0.6 (11/1722) | 3 (0–8) | 18 (8–28) |
| 5–8 | 32 | 2.2 (13/594) | 16 (3–29) | 49 (29–70) |
| 9–12 | 5 | 6.8 (4/59) | 60 (17–100) | 80 (44–100) |
| Total | 123 | 1.0 (28/2888) | 11 (5–16) | 30 (21–39) |
The annual rupture rate is calculated by dividing the number of patients with an aneurysm rupture by the person-years of follow-up, shown in parentheses.
PHASES score groups are shown both without (1st score) and with (2nd score) Finnish population points (5 points), which have no effect on significance levels (p = 0.008, log-rank test for differences in cumulative rupture rates between PHASES score groups). Differences were for < 2 versus 6–7 (p = 0.026) and > 7 (p = 0.006) groups, and for 2–5 versus 6–7 (p = 0.068) and > 7 groups (p = 0.01).
p < 0.001 for differences between present score groups. Differences were for 0 versus 1–4 (p = 0.060), 5–8 (p = 0.001), and 9–12 (p < 0.001); for 1–4 versus 5–8 (p = 0.003) and 9–12 (p < 0.001); and for 5–8 versus 9–12 (p = 0.057).
ROC curves for predicting the long-term rupture risk associated with UIAs. The AUC for the present score was 0.755 (SE 0.050, 95% CI 0.657–0.853, p < 0.001), and the optimal cutoff point was obtained at ≥ 5 versus < 5 points for sensitivity (0.607) and specificity (0.789). The AUC of the PHASES score was 0.674 (SE 0.059, 95% CI 0.558–0.790, p = 0.005), and the optimal cutoff point was obtained at ≥ 6 versus < 6 points for sensitivity (0.500) and specificity (0.811). The difference between the areas was 0.0812 (SE 0.0642, 95% CI −0.0446 to 0.207, Z-statistic = 1.265, p = 0.21).
The cumulative rupture rates with both scoring systems are shown by categories inTable 3, which indicates that those with a PHASES score ≥ 6 or a new score ≥ 5 had higher incidences of UIA rupture than those with lower scores. The cumulative rupture rate curves for these scores are shown inFigs. 2and3.
长期累积associa动脉瘤破裂的风险ted with the PHASES score. For detailed statistics, seeTable 2. The markers on the curves indicate censored events.
长期累积associa动脉瘤破裂的风险ted with the present treatment score. For detailed statistics, seeTable 2. The markers on the curves indicate censored events.
UIA破裂的人力资源根据新的得分was 1.42 per unit (95% CI 1.23–1.65, p < 0.001), and the HR according to the PHASES score was 1.24 per unit (95% CI 1.08–1.42, p = 0.002), irrespective of presence of the population variable in the score. The new score predicted aneurysm rupture better, given that it remained significant (1.38 per unit, 95% CI 1.17–1.62, p < 0.001) after adjustment for the PHASES score (1.08 per unit, 95% CI 0.92–1.26, p = 0.35) in the same model.
Discussion
The PHASES score11obtained for the prediction of aneurysm rupture within < 5 years after diagnosis seems to predict the almost lifelong rupture risk for patients of working age to a moderate degree, but is a somewhat weaker predictor than the score obtained previously from this cohort.12This difference seems mainly to be caused by that fact that the PHASES score did not take into account cigarette smoking, the most important risk factor for SAH.24–26Also, the results of the PHASES study were biased by treatment selection of patients, because the younger patients, smokers, and those with larger UIAs were more likely to be treated at baseline than the others and thus had to be excluded from the natural history studies.
Smoking status during the follow-up has been shown to be a significant time-dependent risk factor for aneurysm rupture, and risk for aneurysm rupture decreases if a patient quits smoking during the follow-up.16Prediction of rupture risk decreases during a very long-term follow-up when variables at baseline are used in the score that concerns particularly the patients’ smoking habit.
Simultaneously with the increasing detection of UIAs, the treatment indications are likely to be decreasing, because both the prevalence of smoking and the incidence of SAH will decrease,7,8and because of the detection of UIAs in older people with other diseases, higher treatment risks, or lower treatment indications.11,27,28Existing aneurysms are also less likely to rupture if smoking decreases further.9,12The cost-effectiveness ratio of UIA treatment is likely to increase. Treatment scores for UIAs and their reliability are essential in order to facilitate treatment decisions.
The recently published treatment score12obtained for this prospective cohort with a nearly lifelong follow-up and a minor treatment selection bias suggested that the need for treatment of UIAs among patients of working age can be assessed quite reliably with only 4 variables obtained at baseline: 1) patient age (< 40 years); 2) cigarette smoking at baseline; 3) aneurysm size (diameter ≥ 7 mm); and 4) aneurysm location in the anterior communicating artery (ACoA) or posterior communicating artery (PCoA), or in the bifurcation of the internal carotid artery (ICA). This new scoring was significantly better than the UIA treatment score (UIATS),12which was based on expert opinion but did not include patient data.6In the UIATS measure, the significance of aneurysm size was nullified, because an increase in size favored both aneurysm repair and conservative follow-up to a similar extent. The UIATS also included several factors that have not been shown to increase the risk of an aneurysm rupture.
A large meta-analysis of individual data for 8382 patients in the course of 29,166 person-years in 6 prospective cohort studies showed that there were few predictors of aneurysm rupture during the mean follow-up of 3.5 years: age (< 40 and > 70 years), hypertension (systolic BP > 140 mm Hg or diastolic BP > 90 mm Hg or use of antihypertensive drugs), prior SAH from another aneurysm (risk reduced in univariable analysis but increased in multivariable analysis), aneurysm diameter and location (ACoA, PCoA, and posterior circulation arteries), and Finnish or Japanese populations.11Current smoking, the most important modifiable risk factor for aneurysmal SAH,24–26was not studied. In the meta-analysis only, the Finnish cohort represented a true natural history of UIA, because the Finnish patients were recruited during a period when UIAs were not treated.9,12The patients in the Finnish cohort were more frequently current smokers, and they were 10 to 20 years younger at baseline than the patients in the other cohorts, which would also explain the higher rupture rate. In other cohorts, the patients who had larger UIAs, were younger, or were smokers were more likely to be treated at baseline, leaving patients with a lower aneurysm rupture risk in the natural history cohorts.27,28The PHASES score can be used principally for those patients left in conservative follow-up for various reasons (small UIAs, advanced age, severe incapacitating diseases, a short life expectancy, or increased treatment risks). The difference between the scores was due to the difference in age scoring and the inclusion of smoking in the recent score. Long-term smoking is more likely to increase the risk of UIA growth that results in rupture.18,20
Location of UIAs in the ACoA and PCoA and the maximum size of the UIA were included in both scoring systems, yielding also the highest score points. These locations are known to carry a higher rupture risk than other aneurysms, and the risk may be even higher than has previously been shown in prospective studies because these aneurysms have also been treated more frequently at baseline than other UIAs.11,27,28
The present treatment score predicted long-term UIA ruptures better than did any of the separate factors alone, so that those with a score of at least 5 points had a mean cumulative rupture risk of 50% at 30 years, and a PHASES risk score ≥ 6 points was associated with a similar long-term cumulative rupture risk. The sensitivities of the new treatment score (≥ 5 vs < 5 points) and the PHASES score (≥ 6 vs < 6 points) to aneurysm rupture were 61% and 44%, respectively, in long-term follow-up. This is undesirable when the target of treatment is to prevent a possibly dangerous aneurysm rupture for life. Decreasing corresponding cutoff points to ≥ 1 versus 0 and ≥ 2 versus < 2 would increase sensitivity to almost 100%, but specificity decreases to 10%–20% (i.e., false-positive rate 80%–90%). This would not be a much better result than that obtained by the toss of a coin. Therefore, patients with scores in the range of 5–12 or PHASES scores ≥ 6 should be treated because their aneurysms have a high long-term rupture risk. Patients with scores 1–4 or PHASES scores 2–5 should be individually evaluated, taking into account both other rupture and treatment-associated factors and factors affecting their remaining lifetime.17Elderly patients with small UIAs are more likely to be left untreated because of the low UIA rupture rate.
The purpose of scoring is to predict future lifelong rupture risk of only diagnosed UIAs. It is not suitable to be used for retrospectively scoring aneurysms that have already ruptured. Of ruptured aneurysms, > 90% are < 10 mm in diameter and 75% are < 7 mm.4,29These aneurysms would yield low scores. Most rupturing small aneurysms probably hemorrhage soon after their appearance because of a high relative growth rate of the lesion. Relative aneurysm growth rate takes into account the initial diameter and has been shown to be correlated more highly with aneurysm rupture than with the aneurysm growth rate itself.20
A major strength and advantage of this study is the simple treatment scoring with only 4 variables, which is easy and simple to use in everyday clinical practice. Scoring is suitable also for other aneurysm populations besides those left for conservative treatment because of different reasons. In individual patients themselves, smoking can affect the scoring and thus treatment indication. The strengths of this study are also the complete and almost lifelong follow-up, the minor treatment selection bias, and the fact that patients were not screened, for example, because of family history or other reasons.9,12Correspondingly, this cohort was considered previously to be of good quality for research purposes and to have only slight sources of bias relative to those used in other prospective UIA studies, representing the only true natural history of UIAs.11Although Finns have been considered to have a higher risk for SAH, its incidence is not higher when the study design with its inclusion and exclusion criteria, the accuracy of diagnosis, and the sex and age distributions of the population are carefully taken into account.8The populations in Nordic countries have similar incidences of SAH.
这项研究的一个限制是相对small sample size, despite the significantly longer follow-up time per patient as compared with other large prospective studies, which, on the other hand, had high treatment selection bias. Patients with a previous SAH or multiple UIAs have not been shown to have elevated risk of aneurysm rupture when confounding factors are taken into account.11Additional controlling for smoking in the PHASES study would have reduced the marginal association between prior SAH and the aneurysm rupture risk, because smoking affects both aneurysm rupture and the occurrence of multiple aneurysms.26,30Posterior circulation UIAs may also have a higher risk of rupture, which was not seen in the present series, probably because of the small number of such cases.
Conclusions
The present new, rapid, simple scoring system is easier and slightly better than the PHASES score for evaluating treatment indications and achieving lifelong prevention of UIA ruptures, but the risk score and need for treatment can also be affected by cessation of smoking. The findings suggest that UIAs in patients of working age can be treated with a better focus and improved cost-effectiveness provided that this is undertaken by experienced physicians capable of achieving good treatment results of aneurysm occlusion.
Acknowledgments
This work was supported in part by research grants to Dr. Juvela from the Paavo Nurmi Foundation. The funding source did not have any involvement in the design and conduct of the study; data collection, analysis, and interpretation; preparation, review, writing, or approval of the manuscript; or decision to submit the manuscript for publication. The author has no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.
Disclosures
The author reports no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.
References
-
1 ↑
JuvelaS.Prehemorrhage risk factors for fatal intracranial aneurysm rupture.Stroke.2003;34(8):1852–1857.
-
2
NieuwkampDJ,SetzLE,AlgraA,et al.Changes in case fatality of aneurysmal subarachnoid haemorrhage over time, according to age, sex, and region: a meta-analysis.Lancet Neurol.2009;8(7):635–642.
-
3
KorjaM,SilventoinenK,LaatikainenT,et al.Cause-specific mortality of 1-year survivors of subarachnoid hemorrhage.Neurology.2013;80(5):481–486.
-
4 ↑
SteinerT,JuvelaS,UnterbergA,et al.European Stroke Organization guidelines for the management of intracranial aneurysms and subarachnoid haemorrhage.Cerebrovasc Dis.2013;35(2):93–112.
-
5
EtminanN,BeseogluK,BarrowDL,et al.Multidisciplinary consensus on assessment of unruptured intracranial aneurysms: proposal of an international research group.Stroke.2014;45(5):1523–1530.
-
6 ↑
EtminanN,BrownRDJr,BeseogluK,et al.The unruptured intracranial aneurysm treatment score: a multidisciplinary consensus.Neurology.2015;85(10):881–889.
-
7 ↑
EtminanN,ChangHS,HackenbergK,et al.Worldwide incidence of aneurysmal subarachnoid hemorrhage according to region, time period, blood pressure, and smoking prevalence in the population: a systematic review and meta-analysis.JAMA Neurol.2019;76(5):588–597.
-
8 ↑
KorjaM,LehtoH,JuvelaS,KaprioJ.Incidence of subarachnoid hemorrhage is decreasing together with decreasing smoking rates.Neurology.2016;87(11):1118–1123.
-
9 ↑
JuvelaS,PoussaK,LehtoH,PorrasM.Natural history of unruptured intracranial aneurysms: a long-term follow-up study.Stroke.2013;44(9):2414–2421.
-
10 ↑
KorjaM,LehtoH,JuvelaS.Lifelong rupture risk of intracranial aneurysms depends on risk factors: a prospective Finnish cohort study.Stroke.2014;45(7):1958–1963.
-
11 ↑
GrevingJP,WermerMJH,BrownRDJr,et al.Development of the PHASES score for prediction of risk of rupture of intracranial aneurysms: a pooled analysis of six prospective cohort studies.Lancet Neurol.2014;13(1):59–66.
-
13 ↑
BackesD,RinkelGJ,LabanKG,et al.Patient- and aneurysm-specific risk factors for intracranial aneurysm growth: a systematic review and meta-analysis.Stroke.2016;47(4):951–957.
-
14
HeiskanenO.Risk of bleeding from unruptured aneurysm in cases with multiple intracranial aneurysms.J Neurosurg.1981;55(4):524–526.
-
15
JuvelaS,PorrasM,HeiskanenO.Natural history of unruptured intracranial aneurysms: a long-term follow-up study.J Neurosurg.1993;79(2):174–182.
-
16 ↑
JuvelaS,PorrasM,PoussaK.Natural history of unruptured intracranial aneurysms: probability of and risk factors for aneurysm rupture.J Neurosurg.2000;93(3):379–387.
-
17 ↑
JuvelaS,LehtoH.Risk factors for all-cause death after diagnosis of unruptured intracranial aneurysms.Neurology.2015;84(5):456–463.
-
18 ↑
JuvelaS,PoussaK,PorrasM.Factors affecting formation and growth of intracranial aneurysms: a long-term follow-up study.Stroke.2001;32(2):485–491.
-
19 ↑
JuvelaS,KorjaM.Intracranial aneurysm parameters for predicting a future subarachnoid hemorrhage: a long-term follow-up study.开云体育app官方网站下载入口.2017;81(3):432–440.
-
20 ↑
JuvelaS.Growth and rupture of unruptured intracranial aneurysms.J Neurosurg.2019;131(3):843–851.
-
22 ↑
HackenbergKAM,AlgraA,SalmanRA,et al.Definition and prioritization of data elements for cohort studies and clinical trials on patients with unruptured intracranial aneurysms: proposal of a multidisciplinary research group.Neurocrit Care.2019;30(suppl 1):87–101.
-
23 ↑
HanleyJA,McNeilBJ.A method of comparing the areas under receiver operating characteristic curves derived from the same cases.Radiology.1983;148(3):839–843.
-
24
AndreasenTH,BartekJJr,AndresenM,et al.Modifiable risk factors for aneurysmal subarachnoid hemorrhage.Stroke.2013;44(12):3607–3612.
-
25
JuvelaS.Alcohol consumption as a risk factor for poor outcome after aneurysmal subarachnoid haemorrhage.BMJ.1992;304(6843):1663–1667.
-
26 ↑
JuvelaS,HillbomM,NumminenH,KoskinenP.Cigarette smoking and alcohol consumption as risk factors for aneurysmal subarachnoid hemorrhage.Stroke.1993;24(5):639–646.
-
27 ↑
WiebersDO,WhisnantJP,HustonJIII,et al.未破裂颅内动脉瘤:自然历史, clinical outcome, and risks of surgical and endovascular treatment.Lancet.2003;362(9378):103–110.
-
28 ↑
MoritaA,KirinoT,HashiK,et al.The natural course of unruptured cerebral aneurysms in a Japanese cohort.N Engl J Med.2012;366(26):2474–2482.
-
29 ↑
RutledgeC,JonzzonS,WinklerEA,et al.Small aneurysms with low PHASES scores account for most subarachnoid hemorrhage cases.World Neurosurg.2020;139:e580–e584.
