Microsurgical treatment of ruptured intracranial aneurysms: An Ethiopian experience

Abstract

Introduction

In this ambispective cohort study, the authors share their experience with surgical outcomes of intracranial aneurysms in a resource-limited setting.

Methods

The study spans a 10-year period (January 2010 to December 2022) at Myungsung Christian Medical Center (MCM) in Addis Ababa, Ethiopia. Patient data, aneurysm characteristics, preoperative, intraoperative, and postoperative factors, along with patient outcomes, were collected using a structured questionnaire. Data analysis employed SPSS version 25, utilizing descriptive statistics and Multivariable logistic regression to identify independent predictors of outcomes. The primary outcome measure was the Glasgow Outcome Scale (GOS) at discharge.

Results

The study included 71 patients with ruptured intracranial aneurysms, with 63 aneurysms clipped in 62 patients. No sex predilection was observed. Aneurysms were predominantly in the anterior cerebral circulation (98.6%), with 60.6% in the anterior communicating artery. Aneurysm size was less than 10 mm in 75.76% of cases. Favorable outcomes were achieved in 83.9% of patients, while 16.10% had unfavorable outcomes. Unfavorable outcomes correlated with a worsening neurological grade assessed by the Glasgow Coma Scale (GCS). The occurrence of delayed vasospasm and hemiparesis showed strong predictive value.

Conclusion

The study demonstrated acceptable mortality and favorable outcomes. Patient outcomes in ruptured cerebral aneurysms were primarily influenced by non-modifiable factors, such as GCS on admission and neurological deficit. GCS exhibited a superior predictive value for outcomes compared to commonly used WFNS and Hunt–Hess scales in patients with intracranial aneurysms.

1. Introduction

Intracranial aneurysms, occurring in 3.6–6%, are often asymptomatic and remain undetected until rupture, causing subarachnoid hemorrhage (SAH) in 85% of cases and resulting in significant mortality. Aneurysmal subarachnoid hemorrhage can lead to devastating neurological complications and mortality can reach up to 50%.^{1, 2, 3,}

Treatment options to reduce complications and mortality shifted globally post-International Subarachnoid Aneurysm Trial (ISAT),⁴ favoring endovascular interventions over microsurgery.⁵ However, in sub-Saharan African countries, limited resources maintain microsurgery as the primary modality.⁶ Despite considerable increases in publications on cerebral aneurysms in Africa, the total number of publications remains low. Forty-one of the 54 African countries did not have publications.⁷ Until 2019; aneurysm clipping was done in 10 patients by Ethiopian neurosurgeon, and expat neurosurgeons. The year 2019 was historical where an Ethiopian vascular neurosurgeon trained in South Korean started to actively treat intracranial aneurysms surgically at Myungsung Christian Medical Center (MCM), an affiliated hospital of Addis Ababa University, the only center providing the service in the country.

2. Material and methods

A hospital-based ambidirectional study covers patients undergoing microsurgical treatment for ruptured intracranial aneurysms at MCM from January 2010 to December 2022. We focused on patients undergoing microsurgical treatment for ruptured intracranial aneurysms. Diagnosis involved a head CT for intracranial hemorrhage and CT angiography for aneurysm detection. Excluded were patients with ruptured vertebrobasilar aneurysms referred for endovascular occlusion or conservatively treated.

2.1. Data analysis and associated factors

Data collection involves patient-related factors (age, sex, comorbidities), aneurysm-related factors (location, size, multiplicity), preoperative factors (Hunt–Hess scale, Glasgow Coma Scale, WFNS scale, Fisher grade), intraoperative factors (aneurysm rupture, temporary arterial occlusion duration), and postoperative factors (cerebral edema, vasospasm). Additionally, the patients were divided into two groups: conscious on admission (H–Hs I, II, III) and unconscious or poor neurological state on admission (H–Hs IV, V). The authors assessed mortality and morbidity separately for those two groups. They defined morbidity as the presence of neurological deficits and/or daily life dependency.

2.2. Outcomes

The patient outcomes were determined using three scales: the Glasgow Outcome Scale (GOS) and the modified Rankin scale (mRS). The outcomes were divided into favorable (4 or 5 GOS points at discharge) and unfavorable (1–3 GOS points at discharge). The authors compared the groups of patients with favorable and unfavorable outcomes. The differences in factors related to the patient, to the aneurysm, preoperative factors, intraoperative factors and factors of postoperative course were examined using χ2 and t tests. Multivariable linear regression identifies associations between various factors (dummy variables) with patients outcomes defined as favorable and unfavorable and assessed in GOS, and mRS. In total, 3 multivariable linear regressions were calculated. A P-value of 0.05 was used as the level of statistical significance. Ethical clearance was obtained from the Departmental Research Ethics Committee, and Statistical Package for Social Sciences (SPSS) version 25.0 is used for data entry and analysis, considering a significance level of 0.05.

Data entry and analysis was carried out using Statistical Package for Social Sciences (SPSS) version 25.0. Ethical clearance was obtained from Departmental Research Ethics Committee.

This study aims to present aneurysm characteristics, surgical outcomes, and predictive values for patients with cerebral aneurysms at MCM, Addis Ababa University's Department of Surgery.

3. Results

The number of patients with ruptured intracranial aneurysm admitted and treated surgically between January 2010 and December 2022 is shown in Fig. 1. This number increased dramatically since 2019 after an Ethiopian vascular neurosurgeon trained in S. Korea started clipping of intracranial aneurysms at MCM.

Fig. 1.

Number of patients with rupture intracranial aneurysms treated surgically,2010–2022.

Of the 78 patients with intracranial aneurysms admitted to the ward/ICU, 71(91%)patients had ruptured aneurysms with median age being 47.00.±13.761(range 17–73 Years), and 38(53.5%) were females, 62 undergone intracranial aneurysm clipping ± wrapping and 9 were not operated (3 patients with Cavernous ICA aneurysm, GCS14-15, referred abroad for endovascular treatment, 3 EVD alone done and another 3 conservatively treated becauene they were unfit for surgery with GCS ≤5). No sex predilection was observed, and patients were in the fourth or fifth decade of life.

Preoperative and postoperative characteristics of the patients with ruptured intracranial aneurysm are presented in Table 1. SAH severity was evaluated clinically (H–Hs, GCS, WFNS) and radiologically by the Fisher grade. The time of maintaining temporary clips was defined as Temporary Arterial Occlusion time (TAO- time). The authors calculated the mean TAO-time and number of patients in which intraoperative aneurysm rupture occurred. The cerebral oedema was defined intraoperatively or postoperatively by radiographic images. CT head or MRI of the brain was used to assess delayed vasospasm.

Table 1.

Characteristics of the patients with ruptured intracranial aneurysm enrolled by the study between January 2010 and December 2022, MCM, A.A, Ethiopia.

Factors	n	Percent
Demographics
Age, median (years)	47,00. ±13.761(range 17–73)
Female Sex (F: M)	38/33(1:1.5)	53.5
Comorbidities	66	93.0
Hypertension	52	73,2.
DM	11	15.5
SAH
Hunt–Hess grade, n = 71
I	8	11.3
II	27	38.0
III	20	28.2
IV	15	21.1
V	1	1.4
WFNS, n-66
I	16	24.2
II	30	45.5
III	7	10.6
IV	10	15.2
V	3	4.5
GCS, n = 66
15	12	18.2
14-13	39	59.1
12-7	6	9.1
6-3	9	13.6
Fisher grade (n = 62)
I	11	17.7
II	18	29.0
III	18	29.0
IV	15	24.2
Intraoperative, n-53
TAO – time, median/Mean (minutes)	4:59/4:30
Intraoperative aneurysm Rupture	7	9.9
Delayed vasospasm	9	14.5
Cerebral oedema	7	9.9

The status, location and size of aneurysms are given in Table 2. Aneurysms were present in the anterior cerebral circulation in 98.6.6% and in the posterior cerebral circulation in 1.4%,of the 70 intracranial aneurysms in the anterior circulation, 43(60,6%) were in the anterior communicating artery, 11 (15.5%) in the middle cerebral artery and another,9(12.7%) were in the internal carotid artery. The average size of the aneurysms was 7.63 ± 3.29 mm with extremes of 1.4 mm and 20.6 mm, the size of the aneurysms was less than 10 mm in 50/66 (75.76%) of cases. Ten(14.1%) patients had multiple aneurysms, and only one patient with multiple aneurysms had multiple clipping.

Table 2.

Characteristics of aneurysms, status, location and size between January 2010 and December 2022, MCM, A.A, Ethiopia.

Factors	N	Percent
Status
Ruptured	64	90.1
Unruptured	7	9.9
Location
MCA	11	15.5
ACom	43	60.6
ACA	7	9.9
ICA	9	12.7
PCom	1	1.4
Size
Small (0–5 mm)	15	21.1
Medium (6–10 mm)	44	62.0
Large (11–25 mm)	11	15.5
Giant (>25 mm)	1	1.4
Multiplicity
Single	61	85.9
Multiple	10	14.1

Majority was admitted in good grades, 49.3% (H–Hs I, II), 69.7% (WFNS, I, II) and 72.3% (GCS, 13–15). Fifty-five of 71(77.5%) patients with intracranial aneurysm were in good or fair neurological state on admission (H–Hs I, II, III) and all of them were conscious on admission, while 16 were in poor neurological state (IV,V), 51/62 of surgically treated patients were in good or fair neurological state on admission, in this group at discharge, 2 patients died, 1 became vegetative, 16 were severely or moderately disabled and 33 patients had good recovery (independent in activities of daily life and demonstrated none of the neurological deficits). From the subgroup of 11/62 patients in poor neurological state (unconscious) on admission (H–Hs IV, V) 3 died. Among 8 survivors, 7 were severely or moderately disabled and only 1 patient was in a good neurological state.See Table 3.

Table 3.

Comparison of patients with favorable and unfavorable outcomes between January 2010 and December 2022, MCM, A.A, Ethiopia.

Factors	Favorable outcome (n = 52)	Unfavorable outcome (n-10)	P-value
Age, mean (years)	46.79 ± 11843	45.30 ± 15.420	0.730
Sex			0.304
•Female	25(53.5)	7(63.6)
•Male	27(46.5)	336.4)
Comorbidities			0.85
•Present	40(68.9))	813.8)
•Absent	9(15.5)	11.72)
Aneurysm size, mean (mm)	7.147 ± 2.4613	10.944 ± 5.4927	0.001a
Aneurysm location			0.398
•ACOM	3454,8)	7(11.3)
•MCA	5(8.06)	1(1.60)
•ACA	9(14.5)	0(0.00))
•ICA	3(4.84)	2(3.22)
•PCOM	1(1.6)	0(90.00)
Clinical grade (Hunt–Hess)			0.014a
•Good (I, II)	28(45.16)	3(4.84)
•Fair (III)	1829.03)	2(3.22 %)
•Poor (IV, V)	6(9.70)	5(8.06)
Fisher grade			0.025a
I	9(22.8%)	2(27.3%)
II	17(29.8%)	1(9.1%)
III	17(31.6%)	1(9.1%)
IV	9(15.8%)	6(54.5%)
TAO – time, median (sec)	4.47 ± 2.18	6.21 ± 2.37	0.104
Intraoperative aneurysm rupture			0.010a
No	49(92.98%)	6(63.6%)
Yes	3(3.50%)	4(36.4%)
Delayed vasospasm			0.004a
No	48(9122%)	5(45.5%)
Yes	4(8.77%)	5(54.5%)

^aValues are statistically significant.

A favorable outcome (GOS 4, 5) was observed in 52 patients (83.9%), and unfavorable outcome (GOS 1–3) in 10 patients (16.10%). In the group of conscious patients on admission, 46/51(90.2%) achieved favorable outcomes, the mortality rate was 2/51(3.9 %) and the morbidity rate 21,6% (11/51). In the group of patients in a poor neurological state, 6/11(54.54%) achieved favorable outcomes, mortality was 3/11(27.2 %) and morbidity was 8/11(72,7%). The overall and operative mortality rates were 14.1% (10/71) and 8.10% (5/62) respectively. See Fig. 2.

Fig. 2.

The outcome and mortality of patients in different neurological states on admission.

4. Comparison of the patients with favorable and unfavorable outcomes

Table 3 shows all the factors included in the comparative analysis. Patient factors such as age, sex and comorbidities didn't differ significantly in both groups. Among the factors related to the aneurysm (location, size), size was the only factor that differed significantly in both groups. Mean aneurysm size in the group with unfavorable outcomes was 10.94 mm and was significantly greater than the mean aneurysm size in the group with the favorable outcomes, which was 7.15 mm (p = 0.001).

Both groups of patients differed significantly in terms of all the preoperative factors: H–Hs and Fisher scale. In the group with unfavorable outcomes, 50 % of the patients were in a severe neurological state on admission (H–Hs IV, V), 20 % in H–Hs III and 36 % in a good neurological state (H–Hs I, II). In the group of patients with favorable outcomes, those values were significantly different (p = 0.014): 11.5 %, 34.62% and 53.85% respectively. Fifty five percent of the patients with unfavorable outcomes had Fisher grade IV; compared to 15.8% of the patients with favorable out-comes (p = 0.025).

TAO-time difference was not statistically significant. Intraoperative aneurysm rupture differed significantly in both groups (p = 0.010); it was more frequent in patients with unfavorable outcomes. In patients with unfavorable outcomes vasospasm occurred significantly compared to patients with favorable outcomes(p = 0.004).

5. The outcomes predictors

Three multivariable linear regressions were performed, separately for GOS, mRS, and outcome determined as favorable/unfavorable (Table 4). The results of the analysis and predictor variables are presented in Table 4. There was a strict correlation between the initial clinical condition of the patient and the outcome. Unfavorable outcome was associated with a worsening neurological grade assessed by GCS (the strongest factor in all analyses; the WFNS was present in only one analysis, Hunt–Hess grade in none) on admission. The age of the patient, ICH and computed tomography (evaluated by Fisher grade) were not predictive factors. The occurrence of delayed vasospasm and hemiparesis had a strong predictive value. Pretreatment size and location of intracranial aneurysms had no predictive values, p = 0.512 and 0.731 respectively.

Table 4.

Outcomes predictor variables (multivariable linear regression); r2 – coefficient of determination; β – regression coefficient between January 2010 and December 2022, MCM, A.A, Ethiopia.

Outcome	Variable	R2	β	p-value
GOS	GCS	0577	0.348	0.0001
	Age		−0.004	0.683
	Hemiparesis		1.469	0.0001
	ICH		−0.327	0.246
	Fisher grade		0.048	0.720
	Chronic comorbidities		−0273	0.361
	Hunt–Hess grade		0.186	0.717
	Delayed Vasospasm		0.821	0.004
	WFNS grade		0.679	0.024
	Intra-operative rupture		1.129	0.004
mRS at 3months	GCS	0.628	−0.488	0.0001
	Age		0.008	0.489
	Hemiparesis		−1.474	0.001
	ICH		−0.252	0.563
	Fisher grade		−0.114	0.551
	Chronic comorbidities		0.410	0.295
	Hunt–Hess grade		0.174	0.617
	Delayed Vasospasm		−1.052	0.012
	WFNS grade		−0.735	0.089
	Intra-operative rupture		0.091	0.046
Outcomes favorable/unfavorable	GCS	0.577	0.348	0.0001
	Age		−0.004	0.683
	Hemiparesis		1.469	0,0001
	ICH		−0.327	0.246
	Fisher grade		0.048	0.720
	Chronic comorbidities		−0.273	0.361
	Hunt–Hess grade		0.186	0.717
	Delayed vasospasm		0.394	0.001
	WFNS grade		0.679	0.240
	Intra-operative rupture		0.129	0.004

6. Discussion

Although there are few literature about intracranial aneurysms in Africa, particularly in sub-Saharan countries, intracranial aneurysms are not uncommon.^8,9 Surgery remains the only therapeutic modality available in several African countries. The number of microsurgically clipped intracranial aneurysm were constantly few, range, 1–4 each year from 2010 to 2018, but this number increased from 6 in 2019 to 25 in 2021 and has remained at that level for subsequent year. The paradigm shift in number of clipped aneurysms was mainly due to the addition of one vascular neurosurgeon to the existing neurosurgeons.

In the present study, the observation of equal gender distribution was contrary to report by Molyneux A and.André Tokpa and others who reported female preponderance.,^{4,10, 11, 12} but in agreement to Etminan N et alwith a 1:1 gender ratio.¹³ such discrepancy might be due to small sample size of the current study.

ISAT reported that 95% of aneurysms were in the anterior cerebral circulation^4,14 similarly, majority of the intracranial aneurysms, 98.6.6% are in the anterior circulation, predominantly on anterior communicating artery. This could be explained by the fact that the aneurysms of the posterior cerebral circulation have been excluded because these aneurysms are easily amenable to endovascular treatment and difficult to manage microsurgically. Thioub et al observed that 73% of operated aneurysms were smaller than 10 mm, André Tokpa also reported average size of aneurysms was 6.5 mm and the sizes of the aneurysms were less than 10 mm in 81.4%.^6,8,13 Similar findings were seen in our study where the average size of the aneurysms was 7.63 mm and aneurysms less than 10 mm were found in 50/66(75.76%) of cases.

Meyer, F.B et al reported that the major clinical manifestation of intracranial aneurysms is subarachnoid haemorrhage, which is the first manifestation in more than 90% of cases.¹⁵ This finding agrees well with our report, aneurysmal rupture with onset of SAH was the most common finding in 71(91.0%) the cases.

Despite the progress made in the surgery of intracranial aneurysms, there is no consensus about the optimal time for performing this surgery. In the 1970s the axiom was that surgery should be delayed for ten or more days from the presenting bleed even in patients who were in a good condition.¹⁶Such delay allowed the initial brain swelling and cerebrovascular instability to subside. Maurice-Williams et al, Yao, Z. et al and other authors^{4,14,17,18, 19, 20} in their study observed that early surgery was superior to late surgery in reducing a poor outcome and death rate for both patients in good or poor condition on admission, in our study clipping of intracranial aneurysms was done post SAH days 10–14,for the following reasons: 1) delayed presentation of the patient and diagnosis, 2) financial constraint 3) shortage of ICU beds, 4) the authors and neurosurgeons in this hospital recommends to operate the patient either within 3 days following the rupture or beyond the 2 weeks.

The favorable outcomes (GOS 4, 5) were achieved in 83.9% of patients, the unfavorable outcomes (GOS 1, 2, 3) were achieved in 16.1% of patients. In agreement to Bogdan Czapiga A et al both groups of patients differed significantly in the preoperative factors (Clinical grade (Hunt–Hess, Fisher grading of severity of SAH), aneurysm size, intraoperative, (Intraoperative aneurysm rupture) and postoperative factor (delayed vasospasm).²¹

Multivariable linear regression revealed preoperative factors (GCS, the presence of neurological deficits on admission) directly affecting the outcome while the factors related to the patient (age, chronic comorbidities) and preoperative factors (SAH severity determined by a Fisher grade, Hunt–Hess grade, ICH and Chronic comorbidities do not affect the outcome. WFNS grade had less influence on the outcome. From the intraoperative factors (aneurysmrupture) strongly affected the outcome whilst the length of TAO does not affect the outcomes. In the postoperative course, the presence of vasospasm had much influence on the outcome. In a study of Deruty el al., the outcome was strongly related to the level of consciousness and the age of the patient with aneurismal SAH, but the relationship between the outcome and age was less marked.¹⁹ Similarly, our study showed that age didn't affect the outcome.

In the current study our finding was in sharp contrast to the analysis of Sandalcioglu et al who revealed that intraoperative aneurysm rupture has no impact on the outcome, neither in the patients in good nor poor initial condition. But in agreement to the finding that poor initial clinical condition (H–Hs IV and V) as well as the initial Fisher grades III and IV were strongly associated with a poor clinical outcome.²⁰

In this study, the strongest predictive factor for outcomes is the neurological condition of the patient on admission ²⁰ and neurologic deficit (the largest absolute value of the beta coefficient in multivariable linear regression) determined by GCS The WFNS had a lower predictive value, while the neurological condition of the patient determined by H–Hs and severity of SAH determined by Fisher grade had no predictive value at all.

Although the Hunt–Hess scale and WFNS are the most widely used subarachnoid hemorrhage grading systems, neither system has achieved universal acceptance and their predictive value regarding the patient outcome is questionable. Oshiro et al concluded that the GCS has equal or greater predictive value regarding outcome after SAH than the currently used grading systems and that it has greater reproducibility across observers²² Teasdale at al. states that the GCS should be used to assess the level of consciousness.²³

As the outcomes in patients with ruptured aneurysm depend mostly on their neurological state on admission, it is essential to avoid SAH. ‘Therefore, it is desirable to identify unruptured aneurysms and treat them before they rupture’ In the study of Krisht’ AF et al, small and medium-sized aneurysms constitute a majority of 64%.²⁴ Dr Bryce Weir estimated that the average size of a ruptured aneurysm is less than 10 mm: small and medium-sized aneurysms rupture before reaching large or giant size.²⁵

The International Cooperative Study on the Timing of Aneurysm Surgery, with 2922 patients operated for treatment of ruptured aneurysms in 68 centers in the world reported a mortality of 14% and Osawa et al from Japan reported a ten-year mortality rate of 12.9 %. In a Study done in Côte d’Ivoire by André Tokpaet al. reported mortality rate of 11% Analyzing our results, we found very similar to these mentioned above and to another in the literature.^{14,26, 27, 28, 29, 30, 31}

7. Conclusion

Our result of treatment of intracranial aneurysm is encouraging and comparable to similar studies with acceptable learning curve when we look the overallpostoperative results. In this ambispective single-center study in the microsurgical procedures of intracranial aneurysms, we had acceptable mortality (7,4%) and favorable outcome (83.4%). The outcomes of patients with cerebral aneurysms depend mainly on non-modifiable factors such as the GCS on admission and neurologicaldeficit. GCS has a better predictive value for outcomes in patients with aneurismal SAH than the commonly usedWFNS and Hunt–Hess scales.

Recommendation

Since most aneurysms are small or medium-sized, the authors recommend the treatment of unruptured aneurysms less than 10 mm as a prevention of SAH. We recommend that good results in the treatment of intracranial aneurysm are quite feasible, despite limited resources.

Limitations and applicability of this study

This is a study conducted on a small number of patients. This may have resulted in multiple biases, particularly with respect to clipping. The outcomes were measured at discharge and 3-month, and cognitive outcomes were not measured. The results of this study can be applied only to centers with a similar expertise in management of these patients.

CRediT authorship contribution statement

Hagos Biluts Mersha: Writing – review & editing, Writing – original draft, Software, Methodology, Formal analysis, Data curation, Conceptualization. Thomas Bogale Megerssa: Project administration, Investigation, Conceptualization.

Declaration of competing interest

The authors disclose that there is not any financial and personal relationships with individuals or organizations that could in any way influence the work that we have submitted.

Abbreviations and acronyms

AAU

Addis Ababa University

CT

scan-Computed tomography

GCS

– Glasgow Coma Scale

GOS

– Glasgow Outcome Scale

H–Hs:

the Hunt–Hess scale

ICU

-Intensive Care Unit

ISAT–

International Subarachnoid Aneurysm Trial

MCM

- Myung Sung Christian Medical Center

mRS

– modified Rankin scale

MRI

–Magnetic resonance imaging

TAO–

temporary arterial occlusion

TASH

– Tikur Anbessa Hospital

WFNS–

World Federation of Neurological Surgeons Grading System