In Re: Histologic and Morphologic Comparison of Experimental Aneurysms with Human Intracranial Aneurysms
========================================================================================================

* WE Stehbens

A fundamental precept in education and science is that errors and misleading statements should not go unchallenged. This commentary aims to draw attention to unwarranted conclusions, methodological deficiency, and misrepresentations of the literature in two recent experimental investigations extolling the elastase infusion model of intracranial aneurysms (1, 2).

Cawley et al (1) produced 15 aneurysms in rabbits using the proximal 0.75 cm of the external carotid artery which was ligated distally. A catheter from the femoral artery inserted 0.5 cm into the sausage-shaped arterial stump was fastened by a cinch tie at the origin of the artery and the “aneurysm” emptied of blood was infused with varying doses of porcine pancreatic elastase for 45 minutes. The cinch tie and catheter were then removed and the femoral artery was ligated. The “aneurysms” were harvested 2 to 12 weeks postoperatively and only six were patent. Histologic loss of the elastic tissue, inflammatory cell infiltration, smooth muscle cell hyperplasia, and intimal fibrosis were found in all “aneurysm” walls, which allegedly were otherwise of normal arterial architecture. This contention could not be confirmed from the low magnification photomicrographs provided. Similar infusion of a segment of rat aorta by Halpern et al (3) resulted in dramatic medial necrosis, loss of both elastin and collagen, and a conspicuous pleomorphic inflammatory cell infiltration, increasing significantly during the 6 days of the experiment. Over this time, a dramatic activation of multiple endogenous proteinases was found, though these may have been associated with resolution and repair of the damage rather than pathogenic activity, causing dilatation of the severely damaged wall.

Cawley et al (1) were not justified in claiming to have produced true arterial saccular aneurysms with degeneration of elastic laminae or that the aneurysm more closely approximated the pathophysiology of naturally occurring cerebral aneurysms. The endothelium in the arterial stumps was allegedly “undisturbed” despite the dissection, two ligatures, elastase infusion, anoxia for 45 minutes, and insertion and withdrawal of the catheter. No attempt was made to confirm the absence of traumatic artifact or to assess the effect of the perfusate on the endothelium. Thrombosis in 60% of their specimens contradicts their allegation of normal endothelium.

Rabbits in which aneurysms were produced bilaterally would be quite unacceptable for long-term studies because of the possible long-term deleterious secondary hemodynamic effects of bilateral internal carotid and femoral artery ligation.

In their second study (2) two experimental models of aneurysms were compared with human intracranial aneurysms. The first model consisted of blind stumps of four ligated extracranial rabbit arteries created using the same elastase infusion technique as previously (1), a procedure far removed from the physiological conditions in which cerebral aneurysms develop and progress. There were no control animals. The severity of cellular and matrix damage incurred was unknown and histologic detail inadequately demonstrated. Such tissue requires more than an allocation of 2 weeks to recover, and the absence of intimal proliferation after such treatment is not surprising. The lesions in no way remotely resemble human cerebral aneurysms. The “thin wall” was actually thicker than that of the internal carotid artery in their photomicrograph and the discussion of thrombus and clot suggests confusion and lack of understanding of the structural differences. Thrombus does not form in a blind cul-de-sac ex vivo in the absence of flow.

Experimental venous pouch aneurysms (the second model) in adult pigs had thick walls, as expected in such a large animal (2). To suggest that gross fibromuscular intimal proliferation had developed and extended throughout the wall within 2 weeks is not believable. This is particularly so in view of considerable inflammatory cell infiltration and a prominent and intact internal elastic lamina and media were observed throughout the pouches. Substantial thrombosis was present in two venous pouch aneurysms, yet patency was alleged to be excellent in all four. The reported extensive mural inflammatory cell infiltration and thrombosis at 2 weeks are suggestive of severe damage, infection, or trauma. Significant mural thickening in venous pouch aneurysms occurs over several months as elastic tissue and muscle progressively disappear (4–7).

Fashioning an aneurysmal sac of a thin-walled vein in any species variously as a lateral, berry or fusiform (or spherical) aneurysm reproduces the hemodynamic conditions found in humans. Using a thin-walled vein of viable tissue is more physiological and valid for long-term experimental observations than their first model (1). The wall would be stronger (ie, more fatigue resistant) than that of weak-walled, early, spontaneous cerebral aneurysms in humans. Though the walls usually heal rapidly postoperatively as did control arteriotomies and phlebotomies in the same animals, aneurysms progressively develop atherosclerosis (histologically [4–7] and ultastructurally [8]), and complications including mural tears, thrombi, and rupture over time (4, 5). Such changes do not represent adaptation but are responses to the degenerative effects of intraaneurysmal hemodynamic stresses.

Models must be suitable for studies of aneurysmal growth and behavior under physiological conditions analogous to those in humans. The authors (2) incorrectly alleged venous pouch aneurysms do not manifest the mural changes of true arterial aneurysms nor are their postoperative changes of only 2 weeks' duration indicative of behavioral changes in aneurysm models. I have never stated that venous pouch aneurysms develop extreme intimal proliferation within 2 weeks or that they progress to near obliteration. The authors also allege I denied enlargement of venous-pouch aneurysms when this is not correct (5).

In the second study (2), their knowledge of the structure of cerebral aneurysms is deficient, and expertise is not acquired from examining nonserial sections of five small sacs. Mural thickness in human aneurysms of the dimensions discussed varies substantially with site and plane of section. It is my opinion, based upon study of serial sections of 154 human cerebral aneurysms (9, 10), that the description and measurements of Abruzzo et al (2) are erroneous and reference 10 provides photomicrographs of 15 human cerebral aneurysms and of other small developing aneurysms. To allege intimal thickening is universally absent in human cerebral aneurysms is also false. A characteristic feature of cerebral aneurysms is intimal proliferation, which is a secondary formation with no media between it and the thinned adventitia. This was the feature that correlated with the medial raphes at branch sites and was used to provide spurious support to the now obsolete “congenital theory” (11). The loss of the internal elastic lamina and underlying media constitutes an “atrophic” lesion of atherosclerosis produced by the enhanced hemodynamic stress (12). The early elastic elements in the compensatory mural thickening (proliferative lesion of atherosclerosis) that follows is intimal proliferation in which the early elastic element soon undergoes progressive degeneration, as occurs in tiny experimental or naturally occurring aneurysmal sacs.

Mural thickness of the berry aneurysm wall at the neck or base differs according to the plane of section, tapering when the fork is cut longitudinally to provide Y-shaped sections, but not when cut in a plane at a right angle to the latter, as seems to be the case in the five human aneurysms of Abruzzo el al (2). Their Figure 5C does not appear to be a recently ruptured sac as alleged. The two large organized thrombi that incompletely filled the sac indicate the primary leakage was not recent, the mural thickness at that site being more than twice the mural thickness at any other location.

Cerebral berry aneurysms in humans develop and enlarge slowly over years (11) and neither of these experimental models (1, 2) reproduces the pathophysiological conditions occurring in the production of the human lesion nor can either be used to deduce their etiology.

The mean of three wall measurements does not represent the mean mural thickness of any aneurysm type. To compare mural thickness of three different aneurysmal types involving three different vessels in three different species and sizes of animals serves no useful purpose, and like other methodological deficiencies, is at variance with sound investigative procedure. Cawley et al and Abruzzo et al (1, 2) published articles with poor design, misrepresentation of the literature, and little cognizance of the scientific method. In addition, my name was spelled incorrectly eight times (text and references) (2). More importantly, the conclusions and generalizations were unwarranted and misleading.

## References

1.  Cawley CM, Dawson RC, Shengelaia G, Bonner G, Barrow DL, Colahan ART. **Arterial saccular aneurysm model in the rabbit.** AJNR Am J Neuroradiol 1996;17:176-1766
    
    

2.  Abruzzo T, Shergelaia GG, Dawson RC, Owens DJ, Cawley CK, Gravanis MB. **Histologic and morphologic comparison of experimental aneurysms with human intracranial aneurysms.** AJNR Am J Neuroradiol 1998;19:1309-1314
    
    [Abstract](http://www.ajnr.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiYWpuciI7czo1OiJyZXNpZCI7czo5OiIxOS83LzEzMDkiO3M6NDoiYXRvbSI7czoyMDoiL2FqbnIvMjEvOS8xNzcwLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

3.  Halpern VJ, Nackman GB, Gandhi RH, et al. **The elastase infusion model of experimental aortic aneurysms: synchrony of induction of endogenous proteinases with matrix destruction and inflammatory cell response.** J Vasc Surg 1994;20:51-60
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=8028089&link_type=MED&atom=%2Fajnr%2F21%2F9%2F1770.atom) 
    
    [Web of Science](http://www.ajnr.org/lookup/external-ref?access_num=A1994NX37900008&link_type=ISI) 

4.  Stehbens WE. **Chronic changes in the walls of experimentally produced aneurysms in sheep.** Surg Gynec Obstet 1979;149:43-48
    
    

5.  Stehbens WE. **Histological changes in chronic experimental aneurysms surgically fashioned in sheep.** Pathology 1997;29:374-379
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=9423218&link_type=MED&atom=%2Fajnr%2F21%2F9%2F1770.atom) 

6.  Stehbens WE. **Chronic changes in experimental saccular and fusiform aneurysms in rabbits.** Arch Pathol Lab Med 1981;105:603-607
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=6895299&link_type=MED&atom=%2Fajnr%2F21%2F9%2F1770.atom) 

7.  Stehbens WE. **Chronic vascular changes in the walls of experimental berry aneurysms of the aortic bifurcation in rabbits.** Stroke 1981;12:643-647
    
    [Abstract/FREE Full Text](http://www.ajnr.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6OToic3Ryb2tlYWhhIjtzOjU6InJlc2lkIjtzOjg6IjEyLzUvNjQzIjtzOjQ6ImF0b20iO3M6MjA6Ii9ham5yLzIxLzkvMTc3MC5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

8.  Stehbens WE. **The ultrastructure of experimental aneurysms in rabbits.** Pathology 1985;17:87-95
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=4000718&link_type=MED&atom=%2Fajnr%2F21%2F9%2F1770.atom) 
    
    [Web of Science](http://www.ajnr.org/lookup/external-ref?access_num=A1985AAS6000019&link_type=ISI) 

9.  Stehbens WE. **Histopathology of cerebral aneurysms.** Arch Neurol 1963;8:272-285
    
    

10. Stehbens WE. **Pathology of the Cerebral Blood Vessels.** St Louis, Missouri: CV Mosby, 1972:351-470
    
    

11. Stehbens WE. **Aneurysms.** In: *Vascular Pathology.* Stehbens WE, Lie JT, Eds. London, England: Chapman & Hall 353-414
    
    

12. Stehbens WE. **Atherosclerosis and degenerative diseases of blood vessels.** In: *Vascular Pathology.* Stehbens WE, Lie JT, Eds. London, England: Chapman & Hall 175-269
    
    

*   Copyright © American Society of Neuroradiology