Late recurrence of subarachnoid hemorrhage and intracranial aneurysms

INTRODUCTION — Aneurysmal subarachnoid hemorrhage (SAH) is often a devastating event. However, therapeutic advances have added to the armamentarium for treating this malignant process. As case fatality rates decline, attention is increasingly turned to the management of long-term complications. One of these is the enduring risk of recurrent SAH, which can occur despite successful endovascular or surgical treatment of the ruptured aneurysm.

This topic discusses the risk of recurrent aneurysm formation and SAH after a patient has been treated for an initial SAH. Other topics address acute aspects of aneurysmal SAH, as well as the management of patients with unruptured intracranial aneurysms, and aneurysm screening in other high-risk populations.

●(See "Aneurysmal subarachnoid hemorrhage: Clinical manifestations and diagnosis".)

●(See "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis".)

●(See "Unruptured intracranial aneurysms".)

●(See "Screening for intracranial aneurysm".)

EPIDEMIOLOGY — Cumulative 8- to 10-year incidences of late rebleeding (more than one year after initial SAH) vary from 0.1 to 3.2 percent [1-5]. The risk of SAH recurrence has been estimated to be 15 to 22 times higher than the expected rate of a first SAH in a healthy age- and sex-matched cohort [2,4].

Independent risk factors for recurrent SAH in one study were current smoking, younger age, and multiple aneurysms at the time of the initial SAH [2]. Hypertension was an additional important risk factor for aneurysm regrowth or de novo aneurysm formation in another retrospective study [6]. Another long-term follow-up study of 59 patients with childhood aneurysms found that current and previous cigarette smoking was a risk factor for recurrent and de novo aneurysm formation [7]. Cigarette smoking and hypertension are also established risk factors for both unruptured intracranial aneurysms and aneurysmal SAH. (See "Aneurysmal subarachnoid hemorrhage: Epidemiology, risk factors, and pathogenesis", section on 'Risk factors' and "Unruptured intracranial aneurysms", section on 'Aneurysm formation'.)

CAUSES — Recurrent SAH may result from recurrence of the treated aneurysm, rupture of another preexisting aneurysm in a patient with multiple aneurysms, or rupture of a de novo aneurysm.

In the International Subarachnoid Aneurysm Trial (ISAT), 33 rebleeds occurred in 1644 patients followed for 10 to 18.5 years after treatment; 17 were from the treated aneurysm, 6 were from a preexisting, untreated aneurysm, and 9 were from new aneurysms (the preexisting status of one aneurysm was unknown) [1,8].

Recurrence of the treated aneurysm — Aneurysms may recur after endovascular or open surgical treatment. Endovascularly treated patients appear to be at somewhat higher risk of rebleeding from the original aneurysm than surgically treated patients [1,3,9]. In systematic reviews of intracranial aneurysms, recurrence after endovascular treatment occurred in 21 to 24 percent [10,11].

●Endovascular treatment – Among the 2108 patients originally treated in ISAT, late retreatment was more frequent after endovascular coiling than after clipping (8.6 versus 0.9 percent) [12]. The mean time to late retreatment after endovascular coiling was 21 months. Recurrent SAH from the originally treated aneurysm occurred most commonly in the endovascular treatment group (10 of 13) [1].

The mechanism of aneurysm recurrence in this setting may be related to compaction of coils over time and/or to aneurysm sac regrowth [13]. One randomized trial found that the use of hydrogel-coated coils (designed to improve packing and stability) was associated with fewer cases of aneurysm recurrence compared with standard bare platinum coils (24 versus 33 percent) [14]. The highest risk for recanalization of a coiled aneurysm appears to be in the first six months and is low after two years [3,15-17].

Most of these aneurysm recurrences are not associated with rupture and SAH; the estimated annual hemorrhage rate after coiling of a ruptured aneurysm is between 0.1 and 3 percent [3,11,15,18-20].

Risk factors for aneurysm recurrence from coil compaction include [9,10,12,17-19,21,22]:

•Larger lumen size (>10 mm)

•Larger aneurysm neck size

•Incomplete occlusion

In one predictive model, the Aneurysm Recanalization Stratification Scale, the chance of aneurysm recanalization is stratified based upon aneurysm size (>10 mm versus ≤10 mm), presence of subarachnoid hemorrhage, presence of intraluminal thrombus, modality of treatment (ie, coils only, stent assistance, or flow diversion), and initial degree of obliteration. Based on these factors, the predicted rate of recurrence can range from 4.9 to 100 percent [21,23].

●Surgical clipping – Recurrence of an aneurysm that was successfully surgically clipped appears to be relatively rare [1-3,24]. In a study of 112 patients (140 clipped aneurysms) who had agreed to undergo cerebral angiography at a mean of nine years after clipping, four aneurysm regrowths (3 percent) were detected [25]. In another study of 610 patients treated with surgical clipping, follow-up computed tomographic angiography (CTA) 2 to 18 years after the index SAH revealed an aneurysm at the clip site in 24 patients (4 percent) [26]. Recurrent SAH attributed to a previously clipped aneurysm is even less common [1-3,15,24,26]. As an example, in a cohort of 752 patients with aneurysmal SAH and successful clipping after a mean follow-up of 8 years, only 4 of the 18 subsequent recurrent SAH (0.5 percent) were associated with a recurrent aneurysm at the clip site [2]. In another series of 711 patients, none of the surgically clipped aneurysms were associated with rerupture over a mean of 4.4 years of follow-up [3].

De novo aneurysm formation — The incidence of de novo aneurysms after surgical clipping or endovascular treatment may be overestimated, in part because aneurysms may be missed at the time of initial hemorrhage [25,27]. As an example, CTA was used to screen 495 patients who had had prior surgical clipping of a ruptured aneurysm at a mean of 8 years previously (range 4 to 14 years). In 87 patients (18 percent), at least one aneurysm was found at a different location than the clip site [27]. Of these 87 patients, the original digital subtraction angiography (DSA) or CTA was available for 51 patients (with 62 aneurysms on follow-up study). Comparison of the original and screening studies revealed that 19 of 62 aneurysms (31 percent) were de novo, and 43 (69 percent) were visible in retrospect. Other cohort studies also find that a significant percentage of "new" aneurysms are present on the original angiogram, when it is available for expert review [2,26].

De novo aneurysm formation probably occurs at a low rate. A meta-analysis of 14,968 total aneurysm patients showed an estimated incidence of 0.3 percent formation of de novo aneurysms per patient-year with no significant difference between patients originally presenting with ruptured versus unruptured aneurysms [28]. In one case series, the five-year cumulative incidence after aneurysm coiling was 0.75 percent [29]. Other studies have reported annual incidence of de novo aneurysm formation of 0.3 to 1.8 percent in patients who have had one aneurysm treated [6,25,30,31]. Multiple aneurysms, smoking, female sex, and older age have been associated with de novo aneurysm formation in some studies [29,32].

In a study of 1601 patients with SAH treated with surgical clipping, the incidence rate of de novo aneurysm formation was 1.42 percent per patient-year [33]. Risk factors for development of a de novo aneurysm included age ≤50 years, family history of aneurysm, and multiplicity at initial SAH. Risk factors for growth or rupture of de novo aneurysms included female sex, shorter interval from initial SAH to detection of a de novo aneurysm, multiplicity at initial SAH, and de novo aneurysm diameter ≥4 mm.

Growth of preexisting aneurysms — Approximately 20 percent of patients with aneurysmal SAH have multiple aneurysms. These may be treated at the same time as the index (ruptured) aneurysm depending on their size, location, and other factors. However, smaller unruptured aneurysms and those less accessible to treatment may be monitored rather than treated.

In one study, serial imaging studies were used to follow 87 patients with 111 unruptured aneurysms; 79 patients had ruptured aneurysms clipped at start of follow-up [34]. Unruptured aneurysms increased in size by ≥1 mm in 45 percent of patients and by ≥3 mm in 36 percent. Cigarette smoking was a risk factor for ≥3 mm aneurysm growth. Other follow-up studies in patients after SAH have also documented a significant rate of aneurysm growth in untreated aneurysms [27]. Recurrent SAH occurred in 1.6 percent of patients per year and was significantly predicted by aneurysm growth. Additional risk factors for aneurysm growth are larger aneurysm size, multiple additional aneurysms, and female sex [29].

FOLLOW-UP EVALUATIONS — There is no consensus on whether and how to screen for new or recurrent aneurysms after SAH [4].

At least two decision models have been used to evaluate the utility of follow-up imaging studies:

●In the first study, outcomes after SAH were modeled using expected outcome and complications rates obtained from a literature review. It was assumed that patients had successful obliteration of all aneurysms by surgical clipping or endovascular coiling after the index SAH [35]. Patients were screened with computed tomographic angiography (CTA). The expected quality-adjusted life years were virtually the same (approximately 8.3 years) for no screening, screening once at five years, and screening every two years, regardless of the initial type of treatment. Screening prevented new episodes of SAH, but the benefit was offset by the cost of increased morbidity from diagnostic tests and preventive treatment. As an example, with screening every two years after coiling, the expected rate of SAH decreased from 1.9 to 0.5 percent and mortality decreased from 0.9 to 0.6 percent, but the disability rate increased from 0.5 to 1.9 percent due to complications from angiography and retreatment.

●In a second study, 610 patients with SAH were screened with CTA 2 to 18 years after surgical clipping, and the results of screening were used as input for a decision analysis [26]. Screening every five years (compared with no screening) prevented nearly half of the SAH recurrences, but life expectancy increased only marginally, and these benefits were offset by a negative impact on quality of life and by increased costs. Screening became cost effective but did not increase quality of life in patients when the risks of aneurysm formation and rupture were doubled, and screening was cost effective and improved quality of life in patients with a 4.5-fold increase in both risks. In addition, screening increased quality of life at acceptable costs in patients with fear for a recurrence.

In the face of limited and conflicting data, it is our opinion that patients require comprehensive follow-up after SAH. Extra vigilance is warranted for patients with risk factors for recurrent SAH and aneurysm regrowth, such as incomplete occlusion at initial treatment, large aneurysm size, multiple aneurysms, hypertension, and cigarette smoking.

●For patients treated with endovascular coiling, we obtain immediate evaluation of the coil mass by angiography during the procedure. Follow-up imaging is usually performed at six months, and, if completely occluded, at two years and five years post-procedure. There is variability in longer-term follow-up in terms of type of imaging (CTA, magnetic resonance angiography [MRA], or angiography) and the interval of imaging. Some form of imaging should be conducted for up to 10 years after treatment.

In addition, we recommend DSA at three to six months for all patients who have undergone coiling, as angiography remains the gold standard [36]; although some data suggest that magnetic resonance angiography (MRA) may be sufficiently accurate for this purpose, it is probably to some extent center specific [37,38].

●For patients treated with surgical clipping of aneurysms, we obtain screening with MRA or CTA at three and six months. Additional angiography is performed only if there are worrisome features on the noninvasive studies.

Further follow-up imaging studies depend on the appearance and size of the treated and any other aneurysms, the presence of risk factors for aneurysm recurrence, and the patient's functional status and individual preferences.

Coil artifacts may interfere with interpretation of CTA in patients treated with coiling, whereas MRA interpretation may be impaired by large artifacts around clipped aneurysms [39,40]. Therefore, CTA is preferred for assessment of patients with clipped aneurysms, and MRA is preferred for patients with coiled aneurysms [41,42]. In one study of 60 patients with 74 coiled aneurysms followed with both angiography and MRA, agreement between the two studies was good. In only four aneurysms was recanalization seen on DSA that was not seen on MRA, with the degree of angiographic recanalization in these patients considered too minor (<3 mm) to indicate further treatment [20,43].

MANAGEMENT — The decision to treat de novo aneurysms or enlarging preexisting aneurysms is based on the same factors used for other unruptured aneurysms. (See "Unruptured intracranial aneurysms", section on 'Management'.)

Treatment of a recurrent, previously treated aneurysm may not be the same as the initial approach and may involve either endovascular coiling, stent-assisted coiling, or flow diversion [44,45]. Surgical clipping is also a possibility depending upon the aneurysm morphology [3,26,46].

Retreatment is not benign. In one case series, 11 percent of recoiling procedures were associated with potentially life-threatening or disabling events, while 2 of 12 repeated surgical procedures resulted in death [3].

The severity of the functional and neurologic morbidity incurred during the index SAH is also a consideration in determining whether intervention is likely to be of overall benefit to the patient's quality of life.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Stroke in adults".)

SUMMARY AND RECOMMENDATIONS

●Epidemiology – Patients who have had an aneurysmal subarachnoid hemorrhage (SAH) have a small but enduring risk of an aneurysmal recurrence and rupture relative to the general population, with a cumulative 10-year incidence of SAH as high as 3 percent. Younger age, cigarette smoking, and hypertension appear to be risk factors for recurrence. (See 'Epidemiology' above.)

●Causes – Recurrent SAH may result from recurrence of the treated aneurysm, de novo aneurysm formation, or rupture of another preexisting aneurysm in a patient with multiple aneurysms. (See 'Causes' above.)

•Recurrence after aneurysm treatment – Aneurysms may recur after endovascular or open surgical treatment. The risk of recurrence is higher in patients treated with endovascular techniques (up to 24 percent) than those treated with surgery (up to 4 percent). Incomplete occlusion and larger (>10 mm) aneurysm size are risk factors for recurrence.

•De novo aneurysms – The incidence of de novo aneurysms after surgical clipping or endovascular coiling appears to be up to 1.8 percent of patients per year. The risk is higher in patients with multiple aneurysms and those who smoke.

•Growth and rupture of preexisting unruptured aneurysms – Patients treated for SAH may harbor other unruptured aneurysms. The risk of SAH from these unruptured aneurysm increased with larger aneurysm size, aneurysm growth, multiple additional aneurysms, and female sex.

●Surveillance imaging after treatment – The frequency and type of neuroimaging follow-up depends on many factors including the treatment (endovascular versus surgery) of the index aneurysm, the presence of risk factors for recurrent aneurysmal formation, the number and size of any additional aneurysms, and the neurologic status and preferences of the patient. (See 'Follow-up evaluations' above.)

•Following endovascular treatment – For most patients who have undergone endovascular treatment of their aneurysm, we typically obtain a cerebral angiogram approximately six months after the initial repair. Long-term follow-up is typically performed for 5 to 10 years, but the type and frequency of imaging varies between centers and countries.

•Following surgical clipping – For most patients who have undergone surgical clipping of a ruptured aneurysm, we typically obtain computed tomographic angiography (CTA) at three to six months.