Pathophysiology of Intracranial Aneurysms: COX-2 Expression, Iron Deposition in Aneurysm Wall, and Correlation With Magnetic Resonance Imaging

Jan Rodemerk¹; Andreas Junker²; M.D.; Bixia Chen¹, M.D.; Daniela Pierscianek¹, M.D.; Philipp Dammann¹, M.D.; Marvin Darkwah Oppong¹, M.D.; Alexander Radbruch⁴, M.D.; Michael Forsting⁴, M.D.; Stefan Maderwald³, PhD; Harald H. Quick³, PhD; Yuan Zhu¹, PhD; Ramazan Jabbarli¹, M.D.; Ulrich Sure¹, M.D.; Karsten H. Wrede¹, M.D.

¹Department of Neurosurgery, University Hospital Essen, Essen, Germany
²Clinic for Neuropathology, University Hospital Essen, Essen, Germany
³Erwin L. Hahn Institute for Magnetic Resonance Imaging, University Duisburg-Essen, Essen, Germany
⁴Department of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Essen, Germany

PMID: 32646326
DOI: 10.1161/STROKEAHA.120.030590

ABSTRACT

Background and Purpose

The pathophysiology of development, growth, and rupture of intracranial aneurysms (IAs) is only partly understood. Cyclooxygenase 2 (COX-2) converts arachidonic acid to prostaglandin H₂ which in turn, is isomerized to prostaglandin E₂. In the human body, COX-2 plays an essential role in inflammatory pathways. This explorative study aimed to investigate COX-2 expression in the wall of IAs and its correlation to image features in clinical (1.0, 1.5, 3.0 Tesla) MRI and ultra-high-field 7 Tesla MRI.

Methods

The study group comprised 40 patients with partly thrombosed saccular IAs. The cohort included 17 ruptured- and 24 unruptured IAs, which had all been treated microsurgically. Formaldehyde-fixed paraffin-embedded samples were immunohistochemically (IHC) stained with a monoclonal antibody against COX-2 (DAKO, Clone: CX-294). We correlated Perls’s Prussian blue staining, MRI images, and clinical data with IHC analyzed using the “Trainable Weka Segmentation” algorithm.

Results

Aneurysm dome size ranged between 2 to 67 millimeters. The proportion of COX-2 positive cells ranged between 3.54% to 85.09%. An upregulated COX-2 expression correlated with increasing IA dome size (p = 0.047). Furthermore, there was a tendency of higher COX-2 expression in most ruptured IAs (p = 0.064). At all field strengths, MRI shows wall hypointensities due to iron deposition correlating with COX-2 expression (p = 0.022).

Conclusion

Iron deposition and COX-2 expression in IAs walls correlate with signal hypointensity in MRI, which might, therefore, serve as a biomarker for IA instability. Furthermore, as COX-2 was also expressed in small unruptured IAs, it could be a potential target for specific medical treatment.

INTRODUCTION

The etiology of saccular and fusiform intracranial artery aneurysms (IAs) is not conclusively substantiated yet. An estimated 2.3% of the population develop an IA.¹ Aneurysmal subarachnoid hemorrhage (aSAH) is a heavy burden for the society as the estimated incidence of aSAH ranges between 6-8 cases per 100.000 people/year and morbidity remains high.²Ruptured IAs cause 1-7 % of all strokes in western countries with a high mortality and morbidity rate.³ Due to improved treatment, the case fatality rate for aSAH has declined from 35.4% in 2000 to 27.2% in 2015.⁴ Nevertheless, as it often occurs in the working population (median age 60 years), it causes a relatively high rate of premature death, comparable with ischemic stroke. It leads to an immense loss of quality of life, employment, and active life years.^{5, 6} Therefore, treatment of rupture-prone IAs prior to bleeding event, is of eminent importance. The current treatment options include microsurgical clipping or endovascular coiling, however, both modalities carry certain treatment-associated risks which should be weighed during treatment decision for unruptured IAs.⁷ Today, treatment stratification follows the ISUIA⁸ recommendations, which is solely based on IA size and location. Due to the increasing number of MRI scans and a higher detection rate of small, innocent IAs, we are in need of better treatment stratification methods to avoid overtreatment. On the other hand, we need to be able to identify small, innocent IAs that are unstable. In search of an in-vivo biomarker for IA instability, several MRI features of IA have been studied.^{9, 10}

A large number of diagnostic markers for IA instability have been analyzed in the last years. Still, only very few seem to have the potential to be a future treatment target.¹¹ One of the most promising ones is COX-2. This key enzyme in the endovascular inflammatory process is also known as prostaglandinsynthase-2 (PGHS-2).¹² It catalyzes the isomerization of the COX product prostaglandin H2 to prostaglandin E2, which is a central mediator of inflammatory processes.¹³ Furthermore, positive feedback occurs when PGE2 activates the endothelial prostaglandin receptor (EP2) initiating the activation of nuclear factor-ĸB, which controls the translation and expression of PTGS2 (The gene encoding for COX-2).¹⁴ This positive feedback loop leads to an increased inflammatory process and amplifies the immigration of lymphatic cells, which weakens the morphological vessel structure and promotes the IA wall instability. Aoki et al. also showed that COX-2 is more abundantly expressed in the vessel wall after hemodynamic stress.¹⁵A single dose ASA (325 mg) per day showed already decreased (mainly in males) rupture incidence.^{16, 17} This explorative study aimed to investigate COX-2 expression in the wall of IA and its correlation to image features in clinical (1.0, 1.5, 3.0 Tesla) MRI and ultra-high-field 7 Tesla MRI.

MATERIAL & METHODS

The data that support the findings of this study are available from the senior author (Karsten.Wrede@uk-essen.de) upon reasonable request.

Study cohort

All cases with aneurysm wall tissue samples harvested during microsurgical clipping for IA between 2009 and 2020 were included in the study. The University of Duisburg-Essen ethical committee authorized the study, and all patients or their relatives provided written informed consent. The study was conducted according to the principles of the Declaration of Helsinki and was Health Insurance Portability and Accountability Act (HIPAA) compliant.

Clinical data collection

Digital patient charts were reviewed for demographic, radiographic, and clinical parameters, and stored in a dedicated database using the program Microsoft-Access. Data included radiological images and reports, and clinical data on pre-existing conditions, medications, comorbidities and other risk factors.

Immunohistochemical COX-2 staining

Aneurysm wall sections (1 µm) were deparaffinized and rehydrated through descending alcohol series and cleaned with distilled water. The sections were then immersed in heated citrate buffer (pH = 6.0) for 15 min in a 700 W Microwave oven and afterwards immersed in 3 % hydrogen peroxide in distilled water to block the endogenous peroxidase activity. The tissues were immunohistochemically (IHC) stained with the anti-COX-2 antibody [1:50 dilution, DAKO, Clone: CX-294]. After incubating overnight (>14 h), the sections were washed in phosphate-buffered saline, incubated with the biotinylated secondary antibody [1:200 dilution, Vector Laboratories; BA-9200], and further supplemented with a Horseradish peroxidase/Strepadividin conjugate. Then, the slides were incubated with 3,3-diaminobenzidine as the substrate for 5 minutes to visualize positively immunostained cells. Finally, the slides were counterstained with hematoxylin. The detailed staining protocol is available on request.

Negative control was used in each staining passage to avoid false-positive results due to unspecific binding of the secondary antibody. Moreover, to make sure that the antibody is detecting the specific protein, we used human adrenal parenchyma as positive control tissue (Fig. 1).

Perls’s prussian blue staining

Aneurysm wall sections (1 µm) were deparaffinized and rehydrated through descending alcohol series and cleaned with distilled water. Afterwards, sections were immersed in potassium ferrocyanide (II) 2% (Morphisto 13306.01000) for 20 minutes, cleaned in distilled water and stained with nuclear fast red for 10 minutes. Finally, sections were drained then with Ethanol and Xylol, and mounted the cover slides with fixative oil.

Histological image analysis

Due to anatomical and operative restrictions during microsurgical clipping, IA samples corresponded to the most abluminal part of the IA dome. Histopathological slides were digitalized with an Aperio Brightfield slide scanner (Leica Microsystems, Wetzlar, Germany) in 20x and 40x magnification. In each slide, five representative areas were selected for further evaluation using the CaseViewer software package (version 2.3 RTM, https://www.3dhistech.com/software-downloads). Semi-quantitative analysis was performed with the FIJI software package (version 2.0.0-rc-69/1.52p, https://imagej.net/Fiji/Downloads) utilizing the machine learning tool for microscopy pixel classification “Trainable Weka Segmentation” (doi:10.1093/bioinformatics/btx180). Every picture (min. 16 million pixels) was divided into three groups using four variables for each group to train the segmentation algorithm to differentiate between COX-2 positive signal, normal tissue area, and nuclei. To quantify the expression, we have calculated the coverage of COX-2 positive cells as a percentage correlative to the whole image area. The mean of COX-2 positive cell coverage for all representative regions was calculated as an estimate of COX-2 expression across the entire aneurysm.

Digital subtraction angiography (DSA), computed tomography (CT), and MRI analysis

Four-vessel DSA was performed using a Philips Allura angiography suite (Philips Healthcare, Best, The Netherlands) capable of 3-dimensional (3D) rotational DSA. Magnetic resonance imaging was acquired with various MR systems ranging from 1 Tesla (T) up to 7 T magnetic field strength (Siemens, GE, Philips, Toshiba) [1T (n = 3); 1.5T (n = 18), 3T (n = 5), and 7T (n = 9)]. Pulse sequences included time-of-flight (TOF) and susceptibility-weighted-imaging (SWI). Two experienced neurovascular surgeons (KHW, BC) and an MD student (JR) evaluated the images using the Horos DICOM viewer (version 3.3.5, https://horosproject.org/). Maximum diameter and neck size were measured in multiplanar reconstructions of TOF MRI data to include thrombosed aneurysm parts and in contrast, enhanced CT-angiography for cases without preoperative MRI. Aneurysm wall signal hypointensities of the abluminal IA part were qualitatively assessed in SWI sequences for correlation with histopathological findings.

Statistical analysis

Statistical analysis was performed utilizing the “R” and “R-Studio” software packages (R version 3.6.2, https://cran.r-project.org/bin/windows/base/; R-Studio version 1.2.5033, https://rstudio.com/products/rstudio/download/) with the following additional packages: GGPlot2, GGExtra, Wesanderson, RODBC, Readxl and fBasic. For categorical variables, we calculated proportions and used a two-sided t-test. For continuous variables, we have calculated the mean, median, range, and standard deviation (SD). The Mann-Whitney U test, Kruskal-Wallis multiple comparison tests, and the Spearman correlation were applied for the complete cohort (p_c), the subgroups of unruptured (p_u), and ruptured (p_r) IAs with p-values labeled accordingly. The significance level α was set to 0.05 and the confidence interval was set to 0.95 %.

RESULTS

Study cohort

The study group comprised 40 patients (12 male) with ruptured (n = 17) and unruptured (n = 24) IAs (Tbl.1). The total number of IAs was 71 as multiple IAs were diagnosed in 16 patients. One patient had two of his six unruptured IAs clipped within 6 months, and thus 41 aneurysm sections were available for IHC. Eight patients had long-time (>6 month) medication with ASA (100 mg/day). IA dome size ranged between 2-67 mm (mean: 17.0, median: 12.5, standard deviation [SD]: ±13.7), and neck size ranged between 2-13 mm (mean: 5.4, median: 5.0, SD: ±2.6). Neck sizes in ruptured IAs were below 7 mm. Ages ranged between 19 and 78 years (mean: 53.5, median: 56, SD: ±13.3).

Risk factors and IA size

There was no correlation between IA size and known risk factors like hypertension (p_c = 0.22), family history (p_c = 0.09), female sex (p_c = 0.09), smoking (p_c = 0.40), and obesity (p_c = 0.97) in our study group. Only regular alcohol consumption or alcohol abuse correlated positively with the IA dome size (p_c = 0.01).

COX-2 expression in the IA wall, clinical data, and radiological data

Expression of COX-2 in the IA wall samples varied between 3.53% and 85.08% (mean: 28.11%, median: 23.33%, SD: ±19.27%). There was no correlation between COX-2 expression and known risk factors like hypertension (p_c = 0.99; p_u = 0.21; p_r = 0.13), smoking (p_c = 0.75; p_u = 0.282; p_r = 0.025), and obesity (p_c = 0.73; p_u = 0.344; p_r = 0.78). Furthermore, regular alcohol consumption or alcohol abuse correlated positively with COX-2 expression and in the subgroup of unruptured IAs (p_c = 0.03; p_u = 0.09; p_r = 0.22). IA dome size correlated positively with COX-2 expression (p = 0.048, Fig. 2A). Overall, there was a strong tendency of higher COX-2 expression in ruptured IA (p = 0.064*, Fig. 2C). P-values marked with an asterisk were calculated excluding IAs larger than 40 mm dome size (n = 3) to correct for outliers. Presence of wall thrombus was more likely in larger IAs (p_c = 0.003; p_u = 0.005; p_r = 0.28) and correlated positively with COX-2 expression (p_c = 0.05; p_u = 0.16; p_r = 0.05). The majority of patients with multiple IAs had smaller aneurysms (p = 0.026*) with less COX-2 expression than in patients with single IAs (p = 0.039*). Furthermore, in the subgroup of patients with long-term 100 mg/d ASA intake only 1 out of 8 aneurysms was ruptured compared to 16 out of 33 aneurysms in the non-ASA subgroup (p = 0.032, Fig. 2D). There was no case with history of systemic inflammatory disease, and preoperative CRP values were within normal range, except in one case with admission 5 days after SAH (Hunt & Hess grade III) and preoperative renal failure.

COX-2 expression in the IA wall, iron deposition and MRI signal hypointensities

Iron deposition in aneurysm wall samples varied between 0.5% and 72.4% (mean: 14.8%, median: 5.9%, SD: 18.2%) and correlated positively with COX-2 expression (p_c = 0.002; p_u = 0.001; p_r = 0.49; Fig. 2B). Wall areas with iron deposition were depicted as signal hypointensities in SWI and TOF MRI sequences (p_c = 0.003; p_u = 0.019) and showed a tendency of higher (p_c = 0.13; p_u = 0.06) COX-2 expression (mean: 34.31%) compared to IAs without signal hypointensities in the wall (mean: 18.64%). There was no substantial difference in detection of signal hypointensities between 1.0T, 1.5T, 3T and 7T MRI, but quantification remained limited to ultra-high field MRI (Fig. 3). An example with low COX-2 expression and low iron deposition is shown in Fig. 4. As expected, most giant IAs had high COX-2 expression and showed high rates of iron deposition (Fig. 5). This is supported by the very strong relationship between IA dome size and iron deposition (p_c = 0.0009; p_u = 0.001; p_r = 0.09)

DISCUSSION

Expression of COX-2 in the IA wall had a large variance in our study and ranged between 3.53% and 85.08% with a mean value of 28.11%. Unfortunately, comparison with published data is difficult as other studies categorized COX-2 expression instead of reporting numeric values. Elevated COX-2 expression in females, patients with family history of SAH or regular alcohol consumption might be explained by a higher susceptibility of the immune system in these patients. It has been shown that ethanol activates the innate immune system by acting as an agonist of Toll-like receptor 4 (TLR4)¹⁸ and several authors suggest a higher COX-2 expression in females although this could only be shown in animal experiments so far.¹⁹

Larger IA dome size and elevated COX-2 expression have been suggested as a marker for IA instability by several authors,^{14, 20} but this is the first study to evaluate the direct correlation between both parameters. The correlation between IA dome size and COX-2 expression in our cohort supports the assumption that elevated COX-2 expression in the IA wall is a sign for instability. Furthermore, the tendency of higher COX-2 expression in ruptured IAs underlines the assumption of COX-2 playing an essential role in the pathophysiology of IA wall stability. Many studies have investigated the connection between the formation of a wall thrombus and IA instability.^21-23 In our study cohort, there was a clear correlation between the elevated COX-2 expression and the presence of a wall thrombus, but as in previous studies it remains unclear whether this is a result or the reason for IA wall instability. Previous studies investigated the effect of ASA on IA rupture rates. These studies speculated that its impact on COX-2 and the downregulated inflammation in the IA wall was responsible for the favorable outcome. Rupture prevention might be achieved by blocking this process with antiplatelet drugs.^{24, 25} An ongoing phase III clinical trial studying the effect of clinical dose of aspirin (100 mg/day) combined with intensive blood pressure control (<120 mm Hg) on the risk of IA rupture or growth in patients with unruptured IA (PROTECT-U, ClinicalTrials.gov identifier: NCT03063541) will probably help us to evaluate the use of these treatments soon. The role of multiple IAs for wall instability remains a topic of ongoing debate.²⁶ The lower COX-2 expression in patients with multiple IAs is contrary to the other findings of our study and is most likely due to a selection bias.

Ollikainen et al.²⁷ showed that remodeling of the IA wall and histological findings suggesting instability are associated with chronic inflammation in histopathology. The absence of an internal elastic lamina, erosion of the luminal endothelium, infiltration of inflammatory cells, apoptosis of smooth muscle cells and the presence of myointimal hyperplasia, fibrosis and thrombus are characteristic for IA wall remodeling. These processes are also associated with neovascularization of the intima, formation of vasorum, and finally with wall degeneration. Frosen et al.²⁸ have also suggested several explanations for iron deposition and macrophage accumulation in unstable IA walls. The origin of interstitial iron is likely due to erythrocytes infiltrating the interstitial space from leaking vessels. This iron is then phagocytized by macrophages that migrate into the IA wall. Perl’s Prussian blue-Staining can reveal these processes in histological samples although macrophage infiltration is also present in most IA walls without iron deposition, supporting the hypothesis that various inflammatory processes are involved in macrophage migration.²¹ A reliable in-vivo biomarker for these processes is still pending, but detection of iron accumulation in the IA wall using susceptibility artifacts in MRI is a promising candidate.^{21, 29-31} Iron deposition in our IA wall samples had a large variation, between 0.5% and 72.4%, and showed a strong positive correlation with COX-2 expression in the IA wall. Furthermore, there was no substantial difference in detection of signal hypointensities between 1.5T, 3T, and 7T MRI, but quantification remained limited to ultra-high field MRI (Fig. 3). Our findings are in line with the literature and support the hypothesis of iron accumulation in the IA wall being an in-vivo biomarker for IA instability. ^{21, 30-32}

Table 1.

Patient-ID	Age	Sex	Dome Size [mm]	Neck Size [mm]	Location	Side	Multiple IAs	Ruptured	Thrombosed	COX-2 [%]	FE [%]	CRP [mg/dl]	MRI	Alcohol	Smoking (Pack /Year)	ASA	Family History	Hypertension
01	61	F	15	04	PICA	R		●		15.33	5.17	1.3	NA	1	0			●
02	56	F	19	06	MCA	L		●		10.24	2.41	0.5	NA	2	36		●	●
03	49	M	20	05	MCA	R		●	●	50.33	46.78	0.5	NA	2	35			●
04	40	F	10	03	MCA	R	●	●	●	27.67	4.23	1	NA	2	14
05	50	M	40	08	MCA	R	●		●	85.08	52.35	0.5	●	2	15	●		●
06	69	F	64	13	ICA	R			●	11.89	48.35	0.5	●	2	0	●		●
07	78	F	09	03	PICA	L		●		21.01	22.96	2.7	NA	1	0			●
08	56	M	25	07	MCA	R			●	12.94	2.16	0.5	●	1	20
09	45	F	09	02	PICA	L		●	●	41.6	2.82	0.5		1	0
10	57	F	08	07	MCA	R	●		●	3.53	1.72	0.5	NA	1	15	●
11	64	F	12	05	MCA	R	●	●		34.5	7.35	2	NA	2	0	●		●
12	53	M	20	05	MCA	R	●		●	50.69	27.62	0.5	●	4	35	●		●
13	56	F	06	03	MCA	R	●	●		30.57	13.66	0.5	NA	1	41		●	●
14	70	F	17	10	MCA	R			●	23.33	5.52	0.5		2	60			●
15	55	M	02	02	ICA	R	●	●		14.62	2.75	0.5	NA	NA	0			●
16	26	M	30	NA	MCA	R		●	●	39.74	28.7	0.8	NA	4	15
17	68	F	17	05	MCA	R	●	●	●	23.33	8.9	0.5		2	0
18	61	F	06	03	MCA	R	●		●	7.51	14.2	0.5	NA	2	40			●
19	39	F	06	04	MCA	R	●	●	●	34.17	1.23	0.5	NA	3	26
20	66	M	12	05	MCA	L				18.09	5.77	0.5	NA	2	75		●	●
21	61	M	67	NA	ACA	L			●	33.27	32.77	0.5	●	2	0			●
22	68	F	07	04	MCA	R	●			4.14	2.7	0.5		0	15	●	●
23	50	F	26	08	MCA	L	●		●	8.95	2.17	0.5	●	2	5
24	44	F	27	05	MCA	L			●	57.97	72.44	0.5	●	2	25			●
25	32	M	NA	NA	MCA	R		●	●	55.26	0.94	0.5	NA	0	17
26	53	F	09	02	PICA	R				40.98	13.96	0.5		2	0
27	24	M	14	06	MCA	L	●			11.56	0.97	1.1		1	10
28	61	F	06	04	MCA	R		●		7.93	5.91	1.3	NA	0	0			●
29	38	F	17	NA	ICA	R		●		63.34	19.24	0.5	NA	2	50			●
30	19	M	27	12	MCA	L			●	46.89	62.52	1.3	●	NA	0
31	48	F	10	04	MCA	L	●			8.58	8.54	0.5		2	9			●
32	56	F	14	08	MCA	L	●		●	9.24	7.48	1.3		1	25			●
33	65	F	20	06	MCA	R			●	21.05	10.97	0.5	●	4	60		●	●
34	50	F	12	04	ACoA	R		●	●	59.94	3.98	0.5	NA	4	33
35	58	F	07	04	MCA	R				17.13	0.5	0.5	NA	NA	30
36	60	F	11	04	MCA	R				23.58	4.4	0.5	NA	1	40
37	37	F	14	07	MCA	R			●	31.36	21.27	0.5	●	0	45	●		●
38	56	F	13	05	MCA	L				16.69	3	0.4	NA	2	10	●
39	63	M	11	06	MCA	L		●		38.03	23.39	21.80*	●	NA	0
40a	66	F	10	10	MCA	L	●			35.8	3.13	0.5		1	45		●	●
40b	66	F	07	03	ACA	L	●			4.73	1.71	0.5		1	45		●	●

Dome Size (maximum diameter in mm), Neck Size (IA neck size in mm); Location (ICA: Arteria carotis interna; MCA: Arteria cerebri media; ACA: Arteria cerebri anterior; ACoA: Arteria cerebri communicans anterior; PICA: Arteria cerebri posterior inferior); COX-2 and FE (coverage of positive cells as a percentage, correlative to the whole image area); CRP (analyzed <24 h before surgery);*The patient was admitted 5 days after SAH (Hunt & Hess grade III) and suffered from preoperative renal failure due to extreme dehydration; Alcohol (0 = never, 1 = rarely, 2 = occasionally, 3 = regularly, 4 = abuse), ASA (daily 100 mg ASA medication over at least 6 month); Family History (for SAH), Hypertension (diagnosed according to American Heart Association guidelines); N/A: data not available.

The size of the study cohort was relatively small (n = 41 IA samples), due to the limited number of cases in which resection of the aneurysm dome was feasible without putting the patient at risk. On the other hand, our study included the largest number of cases compared to previous studies of COX-2 expression in the IA wall.^{14, 16, 20, 31} The number of unruptured IA in our cohort was 1.5 fold higher than in the largest previously published study on COX-2 expression in the IA wall.¹⁴ Furthermore, our cohort is in line with the epidemiological distributions known from larger clinical IA series.^{6, 11, 26, 33, 34} Subgroup analyses showed similar results for most of the parameters and support the proposed pathophysiological mechanisms especially for unruptured IAs. To which extent the differences in ruptured IAs are part of the pathophysiology or just an artefact should be investigated in future studies. In all cases, two parallel clips were placed on the IA neck to achieve complete IA occlusion before the tissue samples were harvested. Only in one case of a ruptured 2 mm IA tissue was harvested distally of the mini-clip. Therefore, all other included IAs had a dome size larger than 6 mm and extrapolation of our data to smaller IAs remains limited. As smaller, stable IA are not accessible for examination one can only evaluate the impact of histological features in larger, unstable IAs. Co-registration of histological findings and MRI remains limited. Nevertheless, tissue samples were harvested from to most abluminal part of the aneurysm dome, allowing a fairly accurate image correlation. Also, other studies focused on IAs with dome sizes ranging between 2 and 35 mm^{20, 30, 31, 35, 36}, whereas several studies are lacking any information on IA dome size.^{14, 16, 37} Beside regular alcohol consumption, there was no relevant correlation between the known risk factors (hypertension, family history, female sex, smoking, obesity) and IA dome size in our study cohort.^{38, 39}

CONCLUSION

Iron deposition and COX-2 expression in the IA wall showed a strong positive correlation with IA dome size. Moreover, ruptured IAs had elevated COX-2 expression compared to unruptured IAs supporting the association of COX-2 expression and IA instability. Wall signal hypointensities in MRI might therefore serve as an in-vivo biomarker. We strongly advocate the establishment of a multicenter registry of histopathological samples and corresponding MRI data to investigate the pathophysiology of aneurysm instability further.

ACKNOWLEDGMENTS

None.

FUNDING

Karsten H. Wrede and Bixia Chen were supported by the University Duisburg Essen (Programm zur internen Forschungsförderung Essen).

DISCLOSURE

Outside the submitted work, Karsten H. Wrede received personal fees from Biogen for expert opinion on aneurysms and vestibular schwannomas, and Alexander Radbruch received grants from Guerbet, Bayer and personal fees from Guerbet, Bayer, and Sanofi.

CONFLICT OF INTEREST

None.

REFERENCES

1.            Vlak MH, Algra A, Brandenburg R, Rinkel GJ. Prevalence of unruptured intracranial aneurysms, with emphasis on sex, age, comorbidity, country, and time period: A systematic review and meta-analysis. Lancet Neurol. 2011;10:626-636

2.            Linn FH, Rinkel GJ, Algra A, van Gijn J. Incidence of subarachnoid hemorrhage: Role of region, year, and rate of computed tomography: A meta-analysis. Stroke. 1996;27:625-629

3.            Feigin VL, Lawes CM, Bennett DA, Anderson CS. Stroke epidemiology: A review of population-based studies of incidence, prevalence, and case-fatality in the late 20th century. Lancet Neurol. 2003;2:43-53

4.            Udy AA, Schweikert S, Anstey J, Anstey M, Cohen J, Flower O, et al. Critical care management of aneurysmal subarachnoid haemorrhage in australia and new zealand: What are we doing, and where to from here? Crit Care Resusc. 2017;19:103-109

5.            Johnston SC, Selvin S, Gress DR. The burden, trends, and demographics of mortality from subarachnoid hemorrhage. Neurology. 1998;50:1413-1418

6.            Nieuwkamp DJ, Setz LE, Algra A, Linn FH, de Rooij NK, Rinkel GJ. Changes in case fatality of aneurysmal subarachnoid haemorrhage over time, according to age, sex, and region: A meta-analysis. Lancet Neurol. 2009;8:635-642

7.            O’Neill AH, Chandra RV, Lai LT. Safety and effectiveness of microsurgical clipping, endovascular coiling, and stent assisted coiling for unruptured anterior communicating artery aneurysms: A systematic analysis of observational studies. J Neurointerv Surg. 2017;9:761-765

8.            Wiebers DO, Whisnant JP, Huston J, 3rd, Meissner I, Brown RD, Jr., Piepgras DG, et al. Unruptured intracranial aneurysms: Natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet. 2003;362:103-110

9.            Cebral JR, Detmer F, Chung BJ, Choque-Velasquez J, Rezai B, Lehto H, et al. Local hemodynamic conditions associated with focal changes in the intracranial aneurysm wall. American Journal of Neuroradiology. 2019;40:510-516

10.          Cebral JR, Vazquez M, Sforza DM, Houzeaux G, Tateshima S, Scrivano E, et al. Analysis of hemodynamics and wall mechanics at sites of cerebral aneurysm rupture. J Neurointerv Surg. 2015;7:530-536

11.          Jabbarli R, Rauschenbach L, Dinger TF, Darkwah Oppong M, Rodemerk J, Pierscianek D, et al. In the wall lies the truth: A systematic review of diagnostic markers in intracranial aneurysms. Brain Pathol. 2020

12.          Simon LS. Role and regulation of cyclooxygenase-2 during inflammation. Am J Med. 1999;106:37S-42S

13.          Dubois RN, Abramson SB, Crofford L, Gupta RA, Simon LS, Van De Putte LB, et al. Cyclooxygenase in biology and disease. FASEB J. 1998;12:1063-1073

14.          Aoki T, Frosen J, Fukuda M, Bando K, Shioi G, Tsuji K, et al. Prostaglandin e2-ep2-nf-kappab signaling in macrophages as a potential therapeutic target for intracranial aneurysms. Sci Signal. 2017;10

15.          Aoki T, Nishimura M, Matsuoka T, Yamamoto K, Furuyashiki T, Kataoka H, et al. Pge(2) -ep(2) signalling in endothelium is activated by haemodynamic stress and induces cerebral aneurysm through an amplifying loop via nf-kappab. Br J Pharmacol. 2011;163:1237-1249

16.          Starke RM, Chalouhi N, Ding D, Hasan DM. Potential role of aspirin in the prevention of aneurysmal subarachnoid hemorrhage. Cerebrovasc Dis. 2015;39:332-342

17.          Chalouhi N, Starke RM, Correa T, Jabbour PM, Zanaty M, Brown RD, Jr., et al. Differential sex response to aspirin in decreasing aneurysm rupture in humans and mice. Hypertension. 2016;68:411-417

18.          Alfonso-Loeches S, Pascual M, Guerri C. Gender differences in alcohol-induced neurotoxicity and brain damage. Toxicology. 2013;311:27-34

19.          St-Louis I, Singh M, Brasseur K, Leblanc V, Parent S, Asselin E. Expression of cox-1 and cox-2 in the endometrium of cyclic, pregnant and in a model of pseudopregnant rats and their regulation by sex steroids. Reprod Biol Endocrinol. 2010;8:103

20.          Hasan D, Hashimoto T, Kung D, Macdonald RL, Winn HR, Heistad D. Upregulation of cyclooxygenase-2 (cox-2) and microsomal prostaglandin e2 synthase-1 (mpges-1) in wall of ruptured human cerebral aneurysms: Preliminary results. Stroke. 2012;43:1964-1967

21.          Sato T, Matsushige T, Chen B, Gembruch O, Dammann P, Jabbarli R, et al. Wall contrast enhancement of thrombosed intracranial aneurysms at 7t mri. AJNR Am J Neuroradiol. 2019;40:1106-1111

22.          Tulamo R, Frosen J, Hernesniemi J, Niemela M. Inflammatory changes in the aneurysm wall: A review. J Neurointerv Surg. 2018;10:i58-i67

23.          Dengler J, Maldaner N, Bijlenga P, Burkhardt JK, Graewe A, Guhl S, et al. Perianeurysmal edema in giant intracranial aneurysms in relation to aneurysm location, size, and partial thrombosis. J Neurosurg. 2015;123:446-452

24.          Hasan DM, Mahaney KB, Brown RD, Jr., Meissner I, Piepgras DG, Huston J, et al. Aspirin as a promising agent for decreasing incidence of cerebral aneurysm rupture. Stroke. 2011;42:3156-3162

25.          Suzuki T, Kamio Y, Makino H, Hokamura K, Kimura T, Yamasaki T, et al. Prevention effect of antiplatelets on aneurysm rupture in a mouse intracranial aneurysm model. Cerebrovasc Dis. 2018;45:180-186

26.          Jabbarli R, Dinger TF, Darkwah Oppong M, Pierscianek D, Dammann P, Wrede KH, et al. Risk factors for and clinical consequences of multiple intracranial aneurysms: A systematic review and meta-analysis. Stroke. 2018;49:848-855

27.          Ollikainen E, Tulamo R, Frosen J, Lehti S, Honkanen P, Hernesniemi J, et al. Mast cells, neovascularization, and microhemorrhages are associated with saccular intracranial artery aneurysm wall remodeling. J Neuropathol Exp Neurol. 2014;73:855-864

28.          Frosen J, Tulamo R, Paetau A, Laaksamo E, Korja M, Laakso A, et al. Saccular intracranial aneurysm: Pathology and mechanisms. Acta Neuropathol. 2012;123:773-786

29.          Matsushige T, Chen B, Ringelstein A, Umutlu L, Forsting M, Quick HH, et al. Giant intracranial aneurysms at 7t mri. AJNR Am J Neuroradiol. 2016;37:636-641

30.          Hasan DM, Mahaney KB, Magnotta VA, Kung DK, Lawton MT, Hashimoto T, et al. Macrophage imaging within human cerebral aneurysms wall using ferumoxytol-enhanced mri: A pilot study. Arterioscler Thromb Vasc Biol. 2012;32:1032-1038

31.          Hasan D, Chalouhi N, Jabbour P, Dumont AS, Kung DK, Magnotta VA, et al. Early change in ferumoxytol-enhanced magnetic resonance imaging signal suggests unstable human cerebral aneurysm: A pilot study. Stroke. 2012;43:3258-3265

32.          Matouk CC, Cord BJ, Yeung J, Malhotra A, Johnson MH, Minja FJ. High-resolution vessel wall magnetic resonance imaging in intracranial aneurysms and brain arteriovenous malformations. Top Magn Reson Imaging. 2016;25:49-55

33.          Molyneux A, Kerr R, Stratton I, Sandercock P, Clarke M, Shrimpton J, et al. International subarachnoid aneurysm trial (isat) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: A randomised trial. Lancet. 2002;360:1267-1274

34.          International Study of Unruptured Intracranial Aneurysms I. Unruptured intracranial aneurysms–risk of rupture and risks of surgical intervention. N Engl J Med. 1998;339:1725-1733

35.          Ishii D, Nakagawa D, Zanaty M, Roa JA, Al Kasab S, Shaban A, et al. Quantitative susceptibility mapping and vessel wall imaging as screening tools to detect microbleed in sentinel headache. J Clin Med. 2020;9

36.          Tulamo R, Frosen J, Paetau A, Seitsonen S, Hernesniemi J, Niemela M, et al. Lack of complement inhibitors in the outer intracranial artery aneurysm wall associates with complement terminal pathway activation. Am J Pathol. 2010;177:3224-3232

37.          Caird J, Napoli C, Taggart C, Farrell M, Bouchier-Hayes D. Matrix metalloproteinases 2 and 9 in human atherosclerotic and non-atherosclerotic cerebral aneurysms. Eur J Neurol. 2006;13:1098-1105

38.          Etminan N, Beseoglu K, Steiger HJ, Hanggi D. The impact of hypertension and nicotine on the size of ruptured intracranial aneurysms. J Neurol Neurosurg Psychiatry. 2011;82:4-7

39.          Backes D, Rinkel GJ, Laban KG, Algra A, Vergouwen MD. Patient- and aneurysm-specific risk factors for intracranial aneurysm growth: A systematic review and meta-analysis. Stroke. 2016;47:951-957