Intracranial Aneurysm in Pediatric Population: A Single-Center Experience

Brain aneurysms affect almost exclusively the adult population [1], and its incidence in children is rare [1‒10], accounting for less than 5% of all aneurysms [1‒3, 5, 7, 8, 11, 12]. Also, compared to adults, they are more complex [4, 10] and differ in location, morphological characteristics, the presentation, and outcome [7, 8].

While the etiology of intracranial aneurysms in adults has widely studied factors in the pediatric population, they still lack certainty [1]. Pediatric intracranial aneurysms (PIAs) are a rare disease that can arise from a variety of different underlying pathological mechanisms [1, 11, 12]. The management of these conditions requires a detailed understanding of the pathology and a multidisciplinary approach.

In general, the clinical presentation of intracranial aneurysms in the pediatric population is due to a mass effect, but the presenting symptoms can be also related to ischemia, subarachnoid hemorrhage (SAH), or, in certain circumstances, a combination of different symptoms [7]. This paper aimed to report a single-center experience with clinical and angiographic aspects of brain aneurysm in children.

We retrospectively reviewed our prospectively maintained database for patients with intracranial aneurysms in our institution from July 2015 to February 2021. Among these, all patients under 18 years of age submitted to diagnostic or therapeutic procedure for an intracranial aneurysm were included.

The aneurysms were classified into saccular or nonsaccular/fusiform based on digital subtraction angiography (DSA) images. For those lesions not located in bifurcations or in the origin of arterial branches, notably those associated with pre- or post-aneurysmatic segmental narrowing and those with mural hematoma visualized on MRI, a dissecting etiology was attributed. All patients were discussed on a case by case basis and treated accordingly.

We collected clinical data during hospitalization and follow-ups during an external consultation or by telephone interviews and imaging follow-up DSA on outpatient request. Children with aneurysms related to arteriovenous malformations were not included in this study.

As the study implied retrospective analysis of anonymized data collected as part of routine clinical care, it did not require formal approval by an ethics committee nor informed written consent by the patient. We informed each parent about their participation in the study. The collected database was declared to the French Data Protection Authority (CNIL).

From July 1, 2015 to February 15, 2021, 12 patients under the age of 18 were submitted to diagnostic or therapeutic procedures in our department. Those children consisted of 7 female and 5 male patients, ranging in age from one-and-a-half months to 16 years (mean age 11.5 years/median 14 years). Three of them had multiple aneurysms, and in total, 17 intracranial aneurysms were assessed in this study (Table 1).

In our cohort of twelve children, none of them had a family history of aneurysm, sickle cell anemia, or any peripheral vascular disease. One child (case 12) had collagen disease, determined by a mutation in the genes COL4A1 and COL4A2, which interfered in the synthesis of type IV collagen.

Five out of twelve children (42%) presented spontaneous SAH due to a ruptured aneurysm. Three patients (25%) had symptoms due to the mass effect of large aneurysms, with compression of cranial nerves or brainstem.

Two patients (17%) had their diagnoses confirmed after imaging workup investigations for headaches. One patient manifested hemiplegia due to nucleocapsular ischemia, caused by an embolic phenomenon from a giant fusiform aneurysm. Yet another patient was diagnosed with cerebral aneurysm after being submitted to stroke endovascular treatment, unrelated to the aneurysm (Table 1).

Of the total of 17 aneurysms, seven (41%) were located in the middle cerebral artery, either in its bifurcation (n = 3) or in the M1 or M2 segments (n = 4). Three aneurysms were in the internal carotid artery (ICA) termination, and another three along the ICA, one in each of the petrous, cavernous, and ophthalmic segments. Four aneurysms (24%) were from the posterior circulation (Table 2).

The maximum diameters of the aneurysms ranged from 1.5 mm to 34 mm (mean 14.2 mm/median 10 mm), with six aneurysms being giant (35%), with diameters greater than 25 mm. Eight (47%) aneurysms were nonsaccular/fusiform and nine were saccular, according to the DSA findings. Two fusiform aneurysms had their etiology related to infections of the central nervous system. Only one of these aneurysms had a mural thrombus image detected by magnetic resonance imaging, namely the one that has spontaneous occlusion.

All eleven symptomatic aneurysms were indicated for endovascular treatment, one of them had spontaneous occlusion, confirmed at the time of the initial diagnostic angiography. The 2 cases of postinfectious nonsaccular/fusiform dissecting aneurysms were treated with aneurysm and parent artery occlusion (PAO) (Fig. 1). Petrous and ophthalmic ICA aneurysms were treated with flow diverter stents. The other four aneurysms were treated with PAO, and two ruptured ICA bifurcation aneurysms were treated with balloon-assisted coiling (Fig. 2). One case of nonsymptomatic aneurysm in the middle cerebral artery bifurcation was submitted to an unsuccessful treatment attempt, being followed up (Table 2). Another nonsymptomatic aneurysm, but with an increase in size in the follow-up period, was submitted to treatment (case 10).

Thus, in the treatment of 12 aneurysms, 4 cases of immediate complications (33%) were reported: two intraprocedural ruptures, one thrombotic phenomenon associated with coil migration, promptly resolved with the capture of the coil and infusion of glycoprotein IIb/IIIa inhibitor and, lastly, the only case of treatment-associated mortality, associated with basilar artery occlusion after treatment with stent, flow diversion, and coils for giant fusiform aneurysm. One case of subacute complication manifested by stroke due to occlusion of the basilar artery was associated with occlusion of dissecting aneurysm and vertebral artery.

Among the 5 cases of treatment-related complications, 1 patient died as a result of the severity of the previous bleeding and the new bleeding suffered during the procedure. Another died due to a posterior fossa stroke. The others had no clinically manifested complications. Another patient, who had undergone occlusion of the middle cerebral artery, had small junctional ischemic infarcts detected on the control MRI, even without clinical manifestations of it.

Three patients underwent retreatment: 1 case of coiling for a ruptured aneurysm of ICA bifurcation that was retreated after 16 months. Another case was of an intracavernous aneurysm that had been initially treated with a flow diverter stent, without occlusion of the lesion, which was submitted to, subsequently, the ICA occlusion. The last case to be submitted to retreatment was the case of treatment of a petrous carotid aneurysm, whose angiographic control performed 11 months after showed ICA occlusion. The patient was submitted to ICA recanalization.

Five aneurysms were not treated. All of them in patients harboring multiple lesions. Four of these aneurysms were considered too small for treatment (mean size 2.0 mm), hence justifying watchful waiting. The other only lesion of considerable size (22 mm) thrombosed spontaneously during the treatment planning period.

Although it has been known since 1971 [13], significant controversy exists as to the exact nature and pathophysiology of pediatric aneurysm formations [13]. Pediatric brain aneurysms can be related to an underlying systemic disease or regional defect, such as connective tissue disorders [6, 12, 13], trauma, infection, abnormal flow, atherosclerosis, familial or developmental syndromes, or vascular segmental vulnerability due to yet unrecognized factors during development [6, 9].

Another theory proposed that childhood saccular aneurysms arise from remnants of small vascular trunks stemming from arterial bifurcations that disappear when the fetal cerebrovascular network matures to form the adult system of major arterial trunks [13]. Such vestigial or incompletely aborted nubbins could become progressively dilated to form aneurysms over time [13].

Most authors agree that both intrinsic defects and acquired insults contribute to the development of aneurysmal disease in children [1, 13, 14], being the damage to the vessel wall the first step in this process [6]. It can be facilitated or induced by any of the above-mentioned conditions, or by other yet unrecognized factors [6]. So, the balance between these conditions and the ability of the vessel wall to repair and compensate for these changes may predict whether an aneurysm forms, grows, ruptures, thromboses, or regresses [13].

Corresponding to less than 5% of all intracranial aneurysms [1, 2, 5, 7], childhood aneurysms are rare and affect boys more than girls [1, 5, 9], notably in individuals above 2 years of age [1]. However, in our series, we found a slightly higher involvement of female patients.

Regarding locations, the authors reported a higher proportion of posterior circulation [1, 4, 8, 9, 15] and ICA terminus [11, 15] aneurysms in children than in adults [1, 3], as well as dissecting aneurysms [3, 15] and giant lesions [3‒5, 7‒10]. In this series, 24% of the lesions (n = 4) were in the posterior circulation, and 18% (n = 3) in ICA terminus. In addition, 35% of the aneurysms (n = 6) were giant, over 25 mm in diameter, reinforcing the idea that pediatric aneurysms are more likely to have larger dimensions.

The highest proportion of giant aneurysms in the pediatric population also refers to the fact that pediatric aneurysms mostly present with nonhemorrhagic symptoms, such as mass effect, headaches, focal neurologic deficits, and seizures [5, 7] with mass effect on nearby structures and so on. In our series, only 42% of the 12 patients (n = 5) had symptoms due to SAH. The others were diagnosed after symptoms resulting from the mass effect, except for 1 patient, who suffered a stroke unrelated to the aneurysm.

Another important point regarding aneurysms in the pediatric population is that the frequency of multiple intradural aneurysms is much lower than in adults [3, 6], with no generally accepted risk factors known for its occurrence [6]. This fact is corroborated by the fact that most PIAs are due to arterial dissections. In this series, among the 12 children evaluated, three had multiple aneurysms (25%). This incidence is similar to the incidence found in adults and diverges from that recommended by the literature for the pediatric population.

On the other hand, in this cohort, the incidence of nonsaccular/fusiform aneurysms (47%, n = 8) and suspected dissections, is greater than that seen in adults, although within the expected for the pediatric population [1, 4, 7]. The dissecting aneurysm group seen in children seems to correspond to a specific type of mural disease [1] and involves the posterior circulation more than the anterior circulation [1, 4]. Focal arterial stenotic segments are often observed proximal or distal to the dissecting aneurysm, suggesting mural damage, as seen in some cases in this series. The impaired vessel wall and associated stenosis can induce spontaneous thrombosis [1, 16], but when treated, the treatment of these aneurysms must also target the mass effect [7].

Approximately, 15% of all pediatric aneurysms are believed to be secondary to an infection [7, 9]. The infectious aneurysms have a peculiar pattern [1, 13]. Those due to bacterial infection are seen more commonly in the distal cortical vessel distribution, whereas infectious aneurysms related to viremia and immune suppression were seen more commonly in the supraclinoid ICA and MCA segments and were often fusiform and multifocal [1]. In this series, one aneurysm was manifested during infection by neurocysticercosis (middle cerebral artery) and another by Propionibacterium acnes (vertebral artery), both with fusiform aspect. Infectious aneurysms will often present with either SAH or intracerebral hemorrhage [7] and whenever possible, such aneurysms have to be treated by occlusion of the parent artery [1].

Guidelines for the treatment of aneurysms are mostly based on adult data and cannot be readily extrapolated to pediatric patients, as children with cerebral aneurysms exhibit important differences from adults in terms of etiology, clinical presentation, aneurysm characteristics, and neurological recovery [3, 15]. Most previous studies revealed higher hospital mortality and complication rates associated with surgical clipping of cerebral aneurysms compared with endovascular coiling [5]. In the study by Alawi et al. [5] it was shown that the mortality of children undergoing treatment for cerebral aneurysm was lower in the endovascular group than in the group undergoing microsurgery (1.65 vs. 6.09%). Also, in this same group of patients, children undergoing surgery had higher rates of hydrocephalus and pulmonary complications. Consistent with adult aneurysm management and the shift toward endovascular concepts, there has been a gradual increase in the use of the endovascular techniques compared to microsurgical clipping since the publication of ISAT [7].

Also, some observational studies have demonstrated the superiority of the endovascular coiling approach in treating aneurysms in children, but most of these studies are small and uncontrolled [5], thus, endovascular treatment is adopted as the first line in many services. Classic saccular or berry aneurysms are rarely seen in neonates, infants, and children under the age of 10; however, their incidence increases with age and up to 20% of aneurysms seen in those below 18 years old may be classified as saccular [7]. Fifty-nine per cent of aneurysms in this series were saccular. Saccular aneurysms affect the anterior circulation more commonly, often located at the classic MCA bifurcation [1, 13] and ICA terminus, such as most saccular aneurysms of this series.

They are relatively easy to discuss due to significant experience of handling similar aneurysms in adults [1]. When they do not have an important mass effect, they can be subjected to coiling, as was done in 2 cases in our cohort.

However, in addition to its exclusion from circulation, the treatment of cerebral aneurysms aims, in the case of lesions with mass effect, to improve the compressive effect. Although it remains a controversial topic, some authors affirm that dense coiling of the aneurysm sac may not relieve the mass effect in case of giant aneurysms [7] and in some cases, parent vessel occlusion or flow diversion is required.

Furthermore, if there is considerable intra-aneurysmal thrombus then the coils can disperse into the thrombus and result in aneurysmal recurrence and the need for repeated treatments. Similarly, if the aneurysm is the consequence of a dissecting process then coil occlusion of the aneurysm will not deal with the underlying pathology. For these reasons both parent vessel occlusion and flow diversion should be considered. These aneurysms can be the most difficult to treat since they can recur even after parent vessel occlusion [7]. In view of this, we performed five treatments by aneurysm and PAO to treat some fusiform and/or large lesions, which also involved the dissecting lesions, of infectious etiology. This treatment modality still represents a safe option for the management of pediatric brain aneurysms.

A therapeutic parent vessel occlusion is often tolerated clinically as the potential for collateral circulation via the circle of Willis and anastomoses or pial collaterals is high in children [1]. In our series, we only witnessed a complication with spontaneous occlusion of the basilar artery in a patient submitted to vertebral artery occlusion. This same patient had significant vasospasm throughout the posterior circulation as a result of the initial SAH.

Dissecting aneurysms are the most difficult cases to manage because of the absence of a neck and the involvement of perforating arteries arising from the dissected aneurysmal wall [1]. In children, there are no generally accepted risk factors for dissections besides previous trauma and being male [6] and there is a theory that states that those dissecting lesions might develop during times of increased intracranial pressure early in development (i.e., during delivery) when herniation of the posterior cerebral artery over the tentorium could cause vessel injury [13], but this is not widely supported.

Anyway, the treatment of these dissecting lesions is done by parent vessel occlusion or by stenting, notably flow diverters. However, the use of flow diverters or intracranial stents in children – as was done to four patients in this series- needs to be considered carefully. The first issue is that of the child’s growth and the expected change in the size of the cerebral vessels might be considered as a contradiction to the use of flow diverters or stents in general [7].

However, it has been shown that the cranial growth is rapid during the early years of life and then slows markedly with virtual complete growth by the age of six, the same age at which the diameter of the intracranial arteries is expected to be almost close to its maximum [7]. So, the 3 cases treated with stents in our series would not be affected by this contraindication, considering that the patients were 12 years old or older.

The second issue is about the need for antiplatelet medication. There are currently no available guidelines on the use of antiplatelet agents in children with cerebrovascular diseases and no definite antiplatelet regimes [7, 10]. This reinforces the close follow-up that should be taken with patients treated with this technique. Even so, in selected cases, flow diversion for treatment of aneurysms in pediatric patients is as efficacious and without increased morbidity as compared to use in adult patients [10].

Concerning the risk of annual recurrence, the series reported by Amelot et al. [2] presented a 2.6% rate, similar to one observed in the adult series of aneurysms, and significantly higher recurrence rate in cases of larger aneurysms [2, 4]. In our series, out of 10 treated cases, three needed to undergo further treatment, only two due to aneurysm occlusion failure. This incidence, although somewhat greater than that described for large series, reflects the complexity of pediatric cases, which often involve giant and fusiform lesions.

PIAs are a rare disease that can arise from a variety of different underlying pathological mechanisms. The management of these conditions requires a detailed understanding of the pathology and a multidisciplinary approach. It should be borne in mind that PIAs have several unique features different from adults, especially its tendency be larger and in not usually locations when compared to adults. Also, despite the availability of new technologies, parent vessel occlusion remains a valid option for aneurysms in the pediatric population.

In addition to representing the single center experience, another limitation of this study is its retrospective design, although all our patients were prospectively registered in dedicated neurovascular databases. PIAs remain rare, and randomized pediatric studies are probably unrealistic. Researchers conducting further studies should help to provide more reliable data as well as a better understanding of this rare but sometimes devastating disease.

The authors express their gratitude toward the patients and their families who accepted to participate in this study. We hope their efforts will benefit future patients.

We submitted our study design to the local Institutional Review Board/Independent Ethics Committee (IRB/IEC) (“Comité de Protection des Personnes”). IRB/IEC did not identify any ethical concerns as an observational study with anonymized data and without any additional therapy or monitoring procedures, Indeed, in accordance with French legislation, the study did not require approval from an ethics committee nor written informed consent from patients since it only involved retrospective analysis of anonymized prospectively collected data as part of routine clinical care; each patient was simply informed of his/her participation in this study and was offered the possibility to withdraw.

The authors have no conflicts of interest to declare.

This study was conducted by the efforts of our department. There was no funding involved.

G.B.A. and A.E. contributed to the conception and design of the study and drafting the manuscript. G.B.A., A.O., A.E., and T.S. contributed to data collection and analysis. J.V.M. and L.S. contributed to revising the manuscript for important intellectual content. All the authors approved the final version to be published.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.