Child's Nervous System

, Volume 26, Issue 10, pp 1319–1327

Microsurgical management of pediatric intracranial aneurysms

Authors

  • Nader Sanai
    • Department of Neurological SurgeryBarrow Neurological Institute
  • Kurtis I. Auguste
    • Department of Neurological SurgeryUniversity of California at San Francisco
    • Department of Neurological SurgeryUniversity of California at San Francisco
Special Annual Issue

DOI: 10.1007/s00381-010-1210-2

Cite this article as:
Sanai, N., Auguste, K.I. & Lawton, M.T. Childs Nerv Syst (2010) 26: 1319. doi:10.1007/s00381-010-1210-2
  • 208 Views

Abstract

Purpose

Pediatric aneuryms are rare and have characteristics that distinguish them from their adult counterparts. There is a greater capacity for pediatric aneurysms to arise de novo and progress rapidly.

Methods

Saccular aneurysms are rarer; fusiform/giant aneurysms are more common. Hemorrhage is less common at presentation than are symptoms that result from mass effect. These patients also present with comorbidities that are unique to children and these conditions may influence treatment selection between minimally invasive procedures and microsurgery. Life expectancy is typically measured in decades for this population and thus treatment durability is of considerable importance.

Results

Our retrospective review indicated that complete aneurysm obliteration occurred in 93% of microsurgery patients versus 79% of endovascular patients. Although functional outcomes were similar for both treatment modalities, the need for additional treatment was over four times more likely in children receiving endovascular therapy.

Conclusion

The need for continued follow-up cannot be overstated for this patient group, therefore, nor can the collaborative efforts of both surgeons and interventionalists to design the most appropriate treatment approach.

Keywords

AneurysmSurgeryEndovascularAngiogram

Introduction

Cerebral aneurysms are believed to form as a result of chronic hemodynamic stress at branch points of arteries. Because the stress points typically require time to manifest aneurysmal changes, aneurysms in children are rare. Estimates of the incidence of pediatric cerebral aneurysms range from 0.5% to 4.6% [7, 25]. To date, approximately 700 cases have been reported in the literature [15, 9, 11, 12, 16, 19, 21, 28, 30, 31, 33, 3537, 39, 40, 4244, 4651, 5356, 58], many of which as case reports [6, 8, 10, 14, 15, 17, 18, 20, 2224, 26, 27, 29, 32, 34, 38, 52]. Some characteristics of aneurysms appearing in this age group have been described in large series [2, 9, 39, 45, 47], including male predominance and preferential location on the internal carotid artery bifurcation. In general, however, the clinical presentations of these aneurysms are as variable as they are infrequent. Some researchers have proposed that these entities are pathologically distinct from adult cerebral aneurysms, thereby requiring specialized treatment [4, 12, 31, 36, 57].

Few reports have asserted that childhood cerebral aneurysms are morphologically different from their saccular counterparts in adults. In our 25-year instructional experience, we have been surprised by the morphological differences from typical berry aneurysms that we observed, with a higher percentage of fusiform aneurysms and a higher percentage of large and giant sizes [13, 41]. In addition, some of these aneurysms were noted to arise de novo, and with rapid progression. These findings suggest that pediatric aneurysms are pathologically distinct, with important treatment implications. Life expectancy and differences in pathology raise important questions with regard to choice of therapy. Specifically, are endovascular techniques as effective as microsurgical techniques in completely obliterating the aneurysm? Is endovascular coiling, which has a rate of recurrence that is documented between 20% and 40%, a realistic therapy when durability of cure over time is in question? Our institutional experience may offer some insight into these and other questions.

Patient presentation

In our experience with 32 patients, aneurysm rupture is not the most common presentation, occurring in just seven patients (22%). Most of the subarachnoid hemorrhage patients have good Hunt–Hess grades and neurological symptoms and signs caused by mass effect from giant aneurysms compressing adjacent cranial nerves or brain structures, the most common being cavernous sinus syndrome. Many children also present with significant medical comorbidities, including polycystic kidney disease, combined immunodeficiency syndrome, cerebral angiodysplasia, congenital hemiatrophy, Schonlein–Henoch purpura, protein C deficiency, collagen vascular disease, growth hormone deficiency, and dwarfism. In our institutional experience, nine patients (28%) had medical comorbidities and six patients (19%) had some history of head trauma.

Aneurysm demographics and treatment paradigms

Over a 25-year interval at UCSF, a total of 43 aneurysms were identified in 32 patients, with eight patients harboring multiple aneurysms (Table 1). Among these eight patients, four had de novo aneurysms that developed after treatment of another aneurysm. Most aneurysms (72%) were located in the anterior circulation, with the middle cerebral artery and cavernous internal carotid artery being the most common sites (Table 2). More aneurysms than expected (28%) were located in the posterior circulation, with the vertebrobasilar confluence and the basilar apex being the most common sites. A disproportionate number of aneurysms were large (nine aneurysms, 21%) or giant (17 aneurysms, 40%). Morphologically, 22 aneurysms (51%) were fusiform or dolichoectatic (Fig. 1), with the remainder having saccular morphology.
Table 1

Summary of all pediatric patients with intracranial aneurysms

Age/Sex

# of aneurysms

Hunt–Hess

Initial treatment

Obliteration

Postoperative deficits

15/F

1

0

Microsurgical

Complete

 

16/M

1

0

Microsurgical

Complete

 

18/M

1

0

Microsurgical

Complete

 

18/M

1

0

Microsurgical

Complete

Weaknessa

18/F

1

0

Endovascular

Complete

 

10/M

1

0

Endovascular

Complete

Weaknessa

2 month/M

1

2

Endovascular

Complete

 

7/F

1

0

Endovascular

Complete

 

7/F

1

0

Endovascular

Complete

Cranial Neuropathya

18/F

1

0

Microsurgical

Complete

 

10/F

3

0

Endovascular

Complete

 

15/F

1

0

Endovascular

Recurrence

 

18/F

1

0

Microsurgical

Complete

 

10/M

1

2

Microsurgical

Complete

 

14/F

2

2

Microsurgical

Complete

 

14/M

1

0

Microsurgical

Complete

Weakness

15/F

3

0

Microsurgical

De novo

 

14/M

1

4

Microsurgical

Complete

 

9/F

1

0

Endovascular

Complete

 

9/M

2

0

Microsurgical

Incomplete

 

10/M

2

0

Endovascular

De novo

 

15/F

2

0

Endovascular

De novo

Weakness

10/M

3

2

Endovascular

De novo

 

2/F

1

0

Endovascular

Complete

 

17/F

1

0

Endovascular

Recurrence

 

9/F

1

0

Microsurgical

Complete

 

17/M

1

1

Endovascular

Residual

 

15/M

2

0

Endovascular

Incomplete

 

15/M

1

3

Endovascular

Recurrence

 

7/F

1

0

Observation

N/A

 

3/F

1

0

Observation

N/A

 

2/F

1

0

Observation

N/A

 

aDenotes transient neurological deficit

Table 2

Locations of intracranial pediatric aneurysms

Location

# of aneurysms (frequency)

Anterior circulation

31 (72.1%)

Internal carotid artery

13 (30.2%)

Middle cerebral artery

11 (25.6%)

Anterior communicating artery

3 (7.0%)

Anterior cerebral artery

2 (4.6%)

Anterior choroidal artery

1 (2.3%)

Pericallosal artery

1 (2.3%)

Posterior circulation

12 (27.9%)

Vertebrobasilar junction

3 (7.0%)

Basilar artery

3 (7.0%)

Posterior cerebral artery

3 (7.0%)

Vertebral artery

2 (4.6%)

Posterior inferior cerebellar artery

1 (2.3%)

https://static-content.springer.com/image/art%3A10.1007%2Fs00381-010-1210-2/MediaObjects/381_2010_1210_Fig1_HTML.gif
Fig. 1

a and b Giant dolichoectatic right middle cerebral artery aneurysm in a 15-year-old girl presenting with seizure. c The exposed aneurysm dome projected inferiorly into the temporal lobe, with lenticulostriates emerging both proximally and distally to the aneurysm, from the afferent and efferent arteries, respectively. d The aneurysm was trapped with temporary proximal and distal clips, allowing for incision and evacuation of the aneurysm, followed by multiple clips to reconstruct the aneurysm neck (adapted from [41])

Management strategies for these 32 patients are summarized in Table 3. Three patients were managed with observation. Two of these patients had dysplastic, fusiform MCA aneurysms that did not enlarge on serial imaging and would otherwise have required bypass and proximal occlusion. The third patient had a non-enlarging pericallosal aneurysm that was detected incidentally. Thirteen patients were treated microsurgically initially with the following techniques: direct clipping (six patients, seven aneurysms), bypass/trapping (two patients, three aneurysms) (Fig. 2), bypass (one patient, two aneurysms), trapping (one patient, one aneurysm), aneurysm excision/reanastomosis (one patient, one aneurysm), proximal parent artery clip occlusion (one patient, one aneurysm), and exploration (one patient). This last patient was immediately referred for endovascular therapy.
Table 3

Summary of microsurgical and endovascular treatment strategies

Treatment strategy

# of patients (# of aneurysms)

Microsurgical

Direct clipping

6 (7)

Bypass/trapping

2 (3)

Bypass

1 (2)

Trapping

1 (1)

Excision/reanastomosis

1 (1)

Proximal parent artery clip occlusion

1 (1)

Exploration

1 (1)

Endovascular

Proximal balloon occlusion

6 (8)

Guglielmi detachable coiling

8 (9)

Balloon embolization

1 (1)

Coiling after distal bypass

1 (1)

Exploration

1 (1)

Observation

3 (3)

https://static-content.springer.com/image/art%3A10.1007%2Fs00381-010-1210-2/MediaObjects/381_2010_1210_Fig2_HTML.gif
Fig. 2

a A giant left cavernous internal carotid artery aneurysm in a 15-year-old boy with progressive diplopia and headache. b The internal carotid artery was exposed and, under EEG burst suppression, occluded with a clip proximal to its entry into the cavernous sinus. c, d, e, and f Following arteriotomy, a saphenous vein graft was stitched into the proximal and distal anastomosis sites. g and h After good flow and hemostasis was confirmed at the bypass graft, a postoperative angiogram demonstrated a patent bypass graft from the petrous internal carotid artery to the supraclinoid internal carotid artery (adapted from [41])

Sixteen patients were treated endovascularly initially with the following techniques: proximal balloon occlusion (six patients, eight aneurysms), Guglielmi detachable coiling (eight patients, nine aneurysms, including one aneurysm in a patient who had his cavernous ICA aneurysm occluded proximally with a balloon), balloon embolization (one patient, one aneurysm), coiling after distal bypass (one patient, one aneurysm), and exploration (one patient). This last patient refused microsurgical treatment.

Patient outcome

Grouping patients by treatment regimens enables comparison of similar cohorts (Table 4). At UCSF, in the group of 13 microsurgical patients, treatment was attempted for 15 aneurysms and accomplished for 14 aneurysms (failure rate, 7%). All but one aneurysm was obliterated completely on postoperative angiography (complete obliteration rate, 93%). In the group of 16 endovascular patients, treatment was attempted for 20 aneurysms and accomplished for 19 aneurysms (failure rate, 5%). The rate of complete obliteration was 79% (15/19).
Table 4

Comparison of microsurgical and endovascular treatment groups

 

Microsurgical

Endovascular

# of patients

13

16

# of aneurysms

15

19

Complete obliteration

14 (93.3%)

15 (78.9%)

Recurrence

0 (0%)

3 (15.8%)

De novo

1 (6.7%)

3 (15.8%)

New neurologic deficits

1 (7.7%)

1 (6.3%)

Mortality

0%

0%

In our experience, no microsurgically treated aneurysms have recurred; whereas three endovascularly treated aneurysms recurred in the same interval; twice in one case (endovascular recurrence rate, 19%). One aneurysm that was balloon embolized required additional balloon embolization, and one coiled aneurysm required additional coil embolization. Another coiled aneurysm was managed microsurgically with bypass and trapping (anterior temporal artery-to-middle cerebral artery side-to-side bypass and trapping of a fusiform, giant MCA aneurysm). De novo aneurysm formation was observed in four patients, one in the microsurgical group (8%) and three in the endovascular group (19%). A new fusiform ACoA aneurysm, formed in a patient whose ICA bifurcation and MCA aneurysms were bypassed and trapped, was treated first with wrapping and later with coiling after it continued to enlarge. In three patients with cavernous carotid aneurysms treated with proximal balloon occlusion, de novo aneurysms formed in the posterior circulation (VBJ in two patients, basilar trunk in one patient). Two of these patients were treated first with saphenous vein bypass grafts from the internal carotid artery to the MCA, followed by clip reconstruction of their vertebrobasilar aneurysms under hypothermic circulatory arrest. The other patient's VBJ aneurysm was left untreated.

One other patient in the endovascular group required additional treatment. His ruptured ICA bifurcation aneurysm was incompletely coiled and he underwent direct clipping of his residual aneurysm the day following his endovascular procedure (Fig. 3). Overall, seven patients required additional treatment, one from the microsurgical group (7%) and six from the endovascular group (32%). Five of these patients switched to the other modality for their additional treatments, with one microsurgical patient requiring endovascular therapy (7%) and four endovascular patients requiring microsurgical therapy (21%).
https://static-content.springer.com/image/art%3A10.1007%2Fs00381-010-1210-2/MediaObjects/381_2010_1210_Fig3_HTML.gif
Fig. 3

a A left internal carotid artery bifurcation aneurysm in a 17-year-old boy who presented with subarachnoid hemorrhage. b Following endovascular coiling, angiography demonstrated residual aneurysm at the base. c Microsurgical exposure of the aneurysm revealed coils within a thin-walled dome. d and e Following clip occlusion, a postoperative angiogram demonstrated complete aneurysm obliteration (adapted from [41])

There were no deaths during the peri-procedural period in the patient series. Two patients had complications that resulted in treatment-related neurological morbidity (2/29, 7%). One patient with an MCA aneurysm underwent an uncomplicated but unsuccessful attempt at microsurgical clipping, had his aneurysm balloon embolized, and suffered a lenticulostriate infarction with new hemiparesis. Another patient with the de novo basilar trunk aneurysm suffered worsened quadriparesis after microsurgical clipping of her aneurysm. Transient neurological worsening was observed in three patients (10%), consisting of a mild hemiparesis in one patient, mild facial weakness in one patient, and multiple new lower cranial nerve deficits in one patient.

Good outcomes (GOS 4 and 5) were observed in all 32 patients, with good recoveries in 25 patients (78%) and moderate disability in seven patients (22%). There was no significant difference in good recovery rates in the microsurgical and endovascular groups (78% and 75%, respectively). Mean duration of follow-up was 5.1 years (range, 1 month–18 years). Extended follow-up was not available in five patients.

Durability of treatment

Lesion recurrence has proven to be a significant concern in the pediatric population, particularly in the setting of vascular disease. For example, AVMs that are completely resected and angiographically obliterated have been shown to recur. In some of these cases, angiogenesis has been postulated and subsequently demonstrated as a likely etiology. Similar concerns arise with aneurysms, particularly for those treated endovascularly. Recurrence has been shown to be a significant concern following coiling, with rates of 20–40% in adults. However, endovascular treatment is growing in popularity, particularly among parents with a strong desire to avoid open surgery for a child. In this setting, the issue of treatment durability is of paramount importance, and thus, we sought to examine aneurysm recurrence after coiling in the pediatric population.

Here we present a large, long-term series of pediatric aneurysms comparing similar groups of patients receiving either microsurgical or endovascular treatment (Table 5). Although the rarity of pediatric aneurysms has prevented any such study to date, the UCSF experience has the advantages of (1) a large cerebrovascular volume, (2) a dedicated team of specialists (vascular neurosurgeons, pediatric neurosurgeons, and neurointerventionalists), (3) long-term clinical and angiographic follow-up, and (4) the latest generations of neurosurgical/neurointerventional technologies. Although our treatment groups were not randomized, the nature and prevalence of the disease makes randomization all but impossible. However, an alternate method of comparing these treatment modalities is to review large case series for each [13, 41].
Table 5

Successful aneurysm obliteration in pediatric aneurysm series with microsurgical and endovascular treatment

 

Total patients

Mean age

Total aneurysms

Microsurgical aneurysm obliteration rate

Endovascular aneurysm obliteration rate

Storrs et al. (47)

29

NR

29

76% (16 of 21)

N/A

Herman et al. (12)

16

8 years

20

94.7% (18 of 19)

N/A

Yazbak et al. (56)

7

9 years

7

100% (7 of 7)

N/A

TerBrugge (50)

21

10 years

25

88.9% (8 of 9)

100% (6 of 6)

Proust et al. (39)

22

13 years

25

95% (19/20)

100% (4/4)

Sanai et al. (41)

32

12 years

43

93.3% (14 of 15)

78.9% (15 of 19)

Successful obliteration defined as no residual aneurysm, no aneurysm recurrence, and no patient mortality after initial treatment

In our institutional experience, both groups of patients presented with aneurysms comparable in number, size, location, and severity. For both treatment groups, there was no operative mortality in any patient. In addition, microsurgical versus endovascular outcomes were similar with respect to the development of new neurologic deficits. Although endovascular patients developed just as many postoperative deficits, there were more incidents of treatment failure and/or aneurysm recurrence (seven endovascular vs. one neurosurgical). One case of aneurysm recurrence following complete endovascular intervention is of particular interest, as the 15-year-old girl's supraclinoid ICA aneurysm was completely obliterated, but found to recur over 2 years later.

A different pathologic entity

The frequency of pediatric aneurysms by location is well-described. In our series, the distribution of aneurysm sacs is consistent with the classic patterns of predominantly anterior circulation involving the cavernous carotid and middle cerebral arteries. Although the frequency of pediatric aneurysms in the posterior circulation remains controversial [39], the 28% seen in our experience is comparable to others. Interestingly, many aneurysms in our series were unique in morphology—frequently giant in size and fusiform in shape. Furthermore, we observed an unusually high incidence of de novo aneurysm formation. These findings raise the possibility that pediatric aneurysms possess a novel pathogenesis as compared with the smaller, berry aneurysms of the adult brain. Consistent with this hypothesis is the high prevalence of medical comorbidities in this population, which may involve developmental vessel wall defects and predispose these patients to aneurysms.

Selection of treatment paradigm

For the pediatric aneurysm patient, the dynamics of selecting a treatment modality are considerably different from that of an adult with a similar lesion. Although concerns for safety and efficacy remain a priority, the issue of durability of treatment takes on new meaning for a child with five to seven decades of life remaining. In our experience, the complication rates, morbidity, and mortality of operative versus endovascular treatment were comparable. With respect to efficacy, both modalities achieved high rates of success, although microsurgical intervention was most successful. For durability, however, the increased recurrence rate demonstrated in the endovascular population, even with a modest follow-up period, warrants serious consideration. Furthermore, the incidence of delayed complications from de novo aneurysm formation was considerably higher in the endovascular patients. These trends raise the question of whether microcatheter manipulation during coiling or balloon occlusion weakens an already defective vessel wall and predisposes it to future aneurysm formation. This remains an area of vascular biology requiring further investigation. In our experience, these negative results from endovascular treatment inevitably led to additional therapy, often in the form of definitive microsurgical treatment. It must be noted, however, that some of these outcomes were the result of older technology that is no longer in use (e.g., balloon embolization), as our current endovascular technology continues to advance.

Although the intrinsic appeal of endovascular therapy to parents must be recognized and respected, a reality reflected in the treatment distribution of this series, parental bias must be tempered by a serious consideration of therapeutic durability, the potential for delayed complications, and the need for additional treatment later in life. By these criteria, microsurgical intervention may be superior to endovascular, as the long life expectancy of pediatric patients requires us to redefine our concept of treatment permanence.

Aneurysm surveillance

Our experience also emphasizes the need for strict patient surveillance among pediatric aneurysm patients. Even after successful and complete treatment, our findings make a strong case for careful follow-up and long-term serial imaging of these patients regardless of treatment modality. Given the pediatric patient's post-treatment lifespan and lifetime risk for not only aneurysm recurrence, but de novo aneurysm formation, the possibility of multimodal therapy must be anticipated. These patients typically require complex techniques for complete resolution, such as bypasses, hypothermic circulatory arrest, and clip reconstruction. Nowhere in neurosurgery is there a clearer need for neurosurgeons and neurointerventionalists to work in specialized teams than with the pediatric aneurysm patient.

Conclusions

Both endovascular and microsurgical treatment of pediatric aneurysms are generally successful modalities. However, our institutional experience suggests that careful patient selection is critical, as microsurgical intervention may yield more durable aneurysm obliteration and represent a definitive treatment for patients at the dawn of life.

Copyright information

© Springer-Verlag 2010