ArticlesDevelopment of the PHASES score for prediction of risk of rupture of intracranial aneurysms: a pooled analysis of six prospective cohort studies
Introduction
Intracranial aneurysms occur in around 3% of the population,1 which means around 15 million inhabitants of the European Union have an unruptured intracranial aneurysm. Owing to the rising availability of brain imaging, the number of incidentally discovered aneurysms is increasing.2 Rupture of intracranial aneurysms results in aneurysmal subarachnoid haemorrhage, a subset of stroke that has high case fatality and morbidity, and occurs at a relatively young age compared with other types of stroke.3, 4, 5 In patients with unruptured aneurysms, the decision whether to treat is often not straightforward. Preventive treatment of intracranial aneurysms carries a risk of combined treatment-related fatality and morbidity of up to 5%.6 Neurosurgical treatment has a higher risk of complications than does endovascular treatment,7 but the risk of rupture after endovascular treatment is slightly higher than after surgery, with annual rupture rates of 0·2% according to a large systematic review.6 The risks of treatment have to be balanced carefully against the risk of rupture.8 However, prediction of the risk of rupture is difficult.
Many prognostic factors for aneurysm rupture have been proposed.9 Risk factors for subarachnoid haemorrhage include aneurysm size and aneurysm site, with higher risks for larger aneurysms and aneurysms in the posterior circulation.10, 11, 12, 13, 14 Multiple aneurysms,12 female sex,9 young age,11, 12 history of subarachnoid haemorrhage,13 and cigarette smoking11 have been suggested as risk factors in some studies, but not in others. Moreover, estimation of absolute risk of aneurysm rupture in a patient based on combination of risk factors is complex and a clinical risk score for aneurysm rupture does not exist. Ideally, one would be able to calculate the risk of aneurysm rupture on the basis of readily available data for patient and aneurysm characteristics.
We undertook a pooled analysis of individual patient data from prospective cohort studies in which data were reported for the natural history of unruptured aneurysms and risk factors predicting rupture. The aim was to establish predictors of aneurysm rupture in patients with unruptured intracranial aneurysms and to provide a risk prediction chart that allows physicians to easily determine the 5-year risk of aneurysm rupture on the basis of a set of routinely assessed patient and aneurysm characteristics.
Section snippets
Search strategy and selection criteria
We did a systematic search in PubMed and Embase, up to July 25, 2013, to retrieve all relevant studies on risk of rupture of unruptured aneurysms. In brief, we used the keywords “(intracranial aneurysm(s) OR cerebral aneurysm(s)) AND (risk of rupture OR aneurysm rupture OR risk factors OR rupture OR unruptured OR subarachnoid hemorrhage) AND (follow-up OR natural history OR natural course)” (appendix). We selected studies that: (1) included 50 or more patients with unruptured intracranial
Results
Table 2 shows the baseline characteristics of the 8382 patients with 10 272 unruptured intracranial aneurysms from the six cohort studies that were included in the pooled analysis. Mean age was 60 years (SD 12) and 68% of the patients were women. Rupture occurred in 230 patients during 29 166 person-years of follow-up (median 2·9 years; range 0–52 years); in 220 patients, a single or the largest aneurysm ruptured. The observed 1-year risk of aneurysm rupture was 1·4% (95% CI 1·1–1·6) and the
Discussion
We have developed a practical risk score (PHASES) that predicts a patient's risk of aneurysm rupture on the basis of a set of routinely assessed patient and aneurysm characteristics. We found that the largest amount of prognostic information was contained in six predictors: age, hypertension, history of subarachnoid haemorrhage, aneurysm size, aneurysm location, and geographical region. Sex, smoking status at time of aneurysm detection, and presence of multiple aneurysms had no important effect
References (29)
- et al.
Prevalence of unruptured intracranial aneurysms, with emphasis on sex, age, comorbidity, country, and time period: a systematic review and meta-analysis
Lancet Neurol
(2011) - et al.
Subarachnoid haemorrhage
Lancet
(2007) - et al.
Changes in case fatality of aneurysmal subarachnoid haemorrhage over time, according to age, sex, and region: a meta-analysis
Lancet Neurol
(2009) - et al.
Missing covariate data in medical research: to impute is better than to ignore
J Clin Epidemiol
(2010) - et al.
Ten-year detection rate of brain arteriovenous malformations in a large, multiethnic, defined population
Stroke
(2010) - et al.
Incidence of subarachnoid haemorrhage: a systematic review with emphasis on region, age, gender and time trends
J Neurol Neurosurg Psychiatry
(2007) - et al.
Endovascular treatment of intracranial unruptured aneurysms: systematic review and meta-analysis of the literature on safety and efficacy
Radiology
(2010) - et al.
Better outcomes with treatment by coiling relative to clipping of unruptured intracranial aneurysms in the United States, 2001–2008
AJNR Am J Neuroradiol
(2011) - et al.
Cost-effectiveness of preventive treatment of intracranial aneurysms: new data and uncertainties
Neurology
(2009) - et al.
Risk of rupture of unruptured intracranial aneurysms in relation to patient and aneurysm characteristics: an updated meta-analysis
Stroke
(2007)