Anatomy of the recurrent artery of Heubner and ischemia of its territory after clipping of aneurysms of the anterior communicating artery

Summary

Aim: The work aims to describe in detail the anatomy of the recurrent artery of Heubner and determine the frequency of ischemia in its territory after clipping of aneurysms of the arteria communicans anterior. Sample and methodology: We performed a laboratory microscopic dissection of the recurrent artery of Heubner as well as literature research of a description of its varieties. In the clinical part, we evaluated post-operative brain CT images of 215 aneurysms of the anterior communicating artery treated with clipping at the Department of Neurosurgery in Ústí nad Labem between 2006–2022. Results: We documented duplication of the arteria recurrens Heubneri, occurring only in 14% of cases, and its extracerebral anastomosis with perforators of the middle cerebral artery, occurring only in 10% of cases. In the clinical part, the frequency of ischemia in the Heubneri recurrent artery territory after surgery for anterior communicating artery aneurysms was found to be 14.4%, for bleeding aneurysms it was 19.6% of cases, and for asymptomatic aneurysms it was of 6% cases. Conclusion: Arteria recurrens Heubneri is one of the arteries which precise anatomical knowledge minimizes the risk of complications when performing vascular and neuro-oncological operations in the area of its course.

Keywords:

aneurysm – ischemia – arteria recurrens Heubneri – arteria communicans anterior

This is an unauthorised machine translation into English made using the DeepL Translate Pro translator. The editors do not guarantee that the content of the article corresponds fully to the original language version.

Introduction

The primary aim of our work was a detailed description of the anatomy of the arteria recurrens Heubneri (recurrent artery of Heubner; RAH), including its varieties, its watershed and embryological development, which we considered as the basis for an introduction to the subject. Furthermore, we decided to retrospectively evaluate our cohort of patients undergoing aneurysm repair of the arteria communicans anterior (AComA) from 2006 to 2022 with the intention of determining the frequency of ischemia in the watershed of this perforating artery, as the RAH is an artery that is at direct risk of manipulation or even clip deployment. We divided our cohort into cases of treatment of bleeding and asymptomatic aneurysms. The available literature assessing the risk of vascular lesion in RAH after AComA clip is not numerous and the results strongly argue in favor of endovascular treatment, so our aim is to assess our series as well so that we can improve the results of microsurgery.

Set and methodology

We started our work by laboratory microscopic dissection on one complete brain with preserved AComA and two hemispheres. In the clinical part of the work we focused on AComA aneurysms treated with clipping at the Neurosurgical Clinic in Ústí nad Labem in the period 2006-2022. 215 patients were evaluated, 83 of these aneurysms were asymptomatic and 132 were bleeding. Our aim was to evaluate the risk of RAH closure related to aneurysm clipping. We assessed the presence of ischemia on follow-up postoperative native brain CT scans performed during early patient care, i.e., 24-72 h after surgery. Images were evaluated in collaboration with a radiologist.

Results

We performed dissection of four RAHs, as documented in Figures 1 and 2. In two cases, we sprayed the vessels with colored latex (1× RAH, 1× arteria cerebri media [ACM]). Despite the very small number of cadaveric material, we were able to document a number of rare variants, the significance and frequency of which we present in the discussion. First, we documented a double fenestration of the AComA (occurring in 0.49%, here in addition in an N-shape with three possible anastomotic channels-Figure 1A). We excavated an extracerebral anastomosis between the RAH and the ACM, occurring in 10% of cases, as well as a common trunk of the RAH with the orbitofrontal artery (Fig. 1B). We documented RAH duplication, occurring in only 14% of cases (Figure 2A, B). In our material, we observed RAH withdrawal from both A2 and A1-A2, but did not encounter its rarer withdrawal from segment A1. In all cases, we observed tortuosity of the RAH course, and in one case, mediolateral overlap of the inlet to the area perforata anterior with perforator from the ACM. Furthermore, we started to document the course of RAH in a targeted photographic manner during vascular and neuro-oncological operations (Figure 3).

In the clinical arm, we identified 31 patients (14.4%) who had postoperative changes of the nature of ischemia in the basin of the arteria recurrens Heubneri (caput ncl. caudati, anteroinferior striatum, anterior arm of the internal capsule) after AComA surgery (fig. 4). RAH ischemia occurred more frequently during clipping of bleeding aneurysms (26 patients), corresponding to 19.6% of the total number of bleeding aneurysms. Five patients (6%) of 83 developed ischemia in the RAH basin after clipping of asymptomatic, i.e. elective, aneurysms. In these five patients, the equivalent of ischemia in the clinical picture was 1× transient hemiparesis, 1× phatic disorder, and 1× delirious state; all three patients had full adjustment to modified Rankin Scale (mRS) level 0, in the fourth patient with pre-existing moderate mental retardation (he was referred for chronic difficulties with blood pressure correction), more fatigue and frustrated central paresis of the mimic muscles were noted at outpatient follow-up, mRS 2, and the patient is unchanged from preoperative status. The ischemia of the fifth patient was also initially asymptomatic.

Discussion

Arteria recurrens Heubneri was named after its discoverer, Johann Otto Leonhard Heubner (1843-1926), a German paediatrician who, apart from being one of the first to successfully administer Behring's and Ehrlich's antitoxin against diphtheria, was also interested in neuroanatomy and described this artery for the first time in 1872. Aitken, a painter at the Massachusetts General Hospital, then referred to it as the Heubner artery in 1909. Subsequently, in 1920, Joseph Shellshear, an Australian anatomist, surgeon, and artillery officer and military medic during World War I, referred to it by the now-used term arteria recurrens Heubneri. However, the path to its naming is not uniform, and it has been named by various authors even in relatively recent times as rostral striatal artery (1953), distal medial striatal artery (1965), telencephalic artery (1984) and long central artery (1986). From the perspective of embryology, the topic of its development and location among other hemispheric arteries has now been clearly treated by Bonasia et al. [1]. The RAH is a cerebral artery appearing very early, arising in the 12mm embryo as one of the plexiform branches of the primitive olfactory artery, the precursor of the arteria cerebri anterior (ACA). Its temporal emergence is preceded only by the arteria choroidea anterior arising in the 9mm embryo. The ACM, on the other hand, arises later, although it is already predominant in the 30mm embryo, but initially it is actually complementary to the RAH. This was already postulated in the first half of the 20th century by comparative phylogenetic work by Abbie [2]. Later, Pierre Lasjaunias [3] used this theory to explain the origin of the rare variant arteria cerebri media accessoria (AccACM), which occurs in its "true form" in only 0.3% of cases and must be distinguished from duplicated ACM. AccACM is actually a hypertrophied RAH supplying cortex and is a remnant of the embryonic plexiform anastomotic plexus. We rely on the review paper by Bonasio et al. [1] for descriptions of the variations in the number and spacing of the RAH, the number of its branches, and the course of , but we will report deviations in individual papers that differ from the averaging. This paper summarizes 20 large cadaveric studies from 1956 to 2018 (total number of hemispheres 2,648). The most robust of these is the study by Matsuda et al [4], describing 714 hemispheres. In summary [1]: the absence of RAH is rare, with only 2% of cases. One RAH is present in 79% of cases, two RAH in 14% and three RAH in 4%. In 90% of cases, the RAH departs within 2 mm of the AComA, most commonly from the A1-A2 junction in 44%, is the first branch of A2 in 43%, and departs from A1 in 14% of cases. Matsuda et al. [4] describe a predominant departure in the 2-mm range of the A1-A2 junction in 72.6% and, interestingly, in three cases (out of 714) they also described an unusual variant of RAH departure 1, from the AccACM 2, from a common trunk with the frontoorbital artery (AOF) -⁠ as we did -⁠ and 3, from a common trunk with the AOF and frontopolar artery. In contrast, a strong predominance of distal spacing from A2 is described by Perlmutter and Rhoton [5], who describe distal spacing as high as 78%. Both Matsuda et al [4] and Loukas et al [6] are sceptical about the occurrence of triple RAH: they report it in 0.14% and 0% of cases, respectively. Uzün et al. [7] even do not find a single case of RAH duplication in their set of 108 hemipheres and report its possible confusion with AOF in other investigators. The diameter of the RAH was measured in 10 studies and the RAH had a mean width of 0.68 mm (0.45-1 mm). The average length of the RAH was also measured in 10 studies (these are not always identical papers to previous ones) and is reported to be 22.9 mm (17-30 mm). Superior position to A1 (I) is present in 38%, anterior position to A1 (II) in 53% and posterior position to A1 (III) in 9%. This important division for the surgeon according to the position of the RAH with respect to A1 into three types was introduced by Gomez at al. [8]; however, their observed frequencies of I 63%, II 34% and III 9% in 60 hemispheres do not correspond to the average of all the studies evaluated -⁠ Bonasia et al. [1] and similarly Matsuda et al. [4] report the most frequent anterior rather than superior position in 62.2%. The study by Perlmutter and Rhoton [5] (100 hemispheres) does not record posterior position (III) at all. The majority of RAH branches including the terminal branches are perforating arteries, Rosner and Rhoton [9] reported that two-thirds of the perforating branches enter the substantia perforata anterior to lateral to the tractus olfactorius, and in 30-40% of cases the extracerebral course of the RAH extends beyond the limen insulae [5]. Branching and anastomoses by selective spraying of ACM and RAH with blue or green and red acrylic and microscopic observation have been extensively studied by Paweł Maga [10-12], who examined 140 hemispheres and performed aesthetic corrosive preparations after vessel filling with resin or polyvinyl chloride. According to him, the number of branches protruding from the RAH ranged from 2 to 30 (mean number 9.4), in contrast to earlier works which indicated that the RAH mainly had the function of perforating arteries. The RAH branches supply the caput nuclei caudati and the anterior arm of the internal capsule in the classical description (Fig. 5), but its watershed can also be extended to the anteroinferior striatum (Broc's diagonal stripe, nucleus basalis Meynerti, nucleus accumbens), the anterior third of the putamen and anterior part of the external pallidum, the genu of the internal capsule, the anterior thalamic nuclei, and the anterior hypothalamus -⁠ these are the perforating branches, but also the areas of the superficial arterioles for the olfactory area, the posterior basofrontal cortex incl. The posterior part of the gyrus rectus, limen insulae and fasciculus uncinatus and, according to Magy's microanatomical dissections, the RAH also gives off perforators to the optic tract in 15.9% of hemispheres [10]. Furthermore, Polish authors describe "connecting branches" -⁠ extracerebral anastomoses mainly with medial lenticulostriate perforators from the ACM (but in one case also with the arteria communicans posterior and 2 times directly with the ACM trunk) in 15.2% [11], and even intracerebral anastomoses in 6.5% [12]. Thus, the extent of damage to target brain tissue structures after RAH closure is then variable or can be compensated for due to the possible presence of these adhesions, which are remnants of Abbie's plexiform anastomotic network. In the extracerebral course of RAH Maga et al. even describe an inferior position to A1, i.e. Gomez type IV in 4%, but otherwise they are in very good agreement with Gomez's work [8] regarding the frequency of occurrence of position I-III, the superior position being also the most frequent according to them [12]. To conclude the anatomical section, we cannot fail to highlight the work of the authors from the Anatomical Institute of the Comenius University Faculty of Medicine in Bratislava (El Falougy H. et al.), who examined 366 hemispheres and used a sophisticated method to measure RAH diameters by calculating the number of pixels in a calibrated digital image and converting them to a metric scale [13]. The average outer diameter of the RAH was 0.6 mm (0.19-1.34 mm), but the authors describe the appropriateness of adding 5-10% to obtain a realistic RAH diameter, as the measurements were performed on formalin-fixed brains. In their work, these authors reported a relatively high absence of RAH in 5.19% of hemispheres.

We compared our clinical results in Ústí nad Labem with the work of radiologists from Stanford University in 2017, who in their numerically balanced retrospective (50 coiled + 50 clipped patients) study presented the results of coiling as superior to those of neurosurgeons [14]. Although patients in the coiled group were on average older and in worse clinical condition (Hunt Hess > 2 in 60% versus 34%), patients in this group were less dependent on nursing care at discharge (mRS < 2) compared with patients after clipping (odds ratio, 3.4; 95% confidence interval, 1.2-12.3; p = 0.04). However, after accounting for patients who died before discharge (three clipped, 6%, and six coiled, 14%), there was no statistically significant difference in clinical outcome between the two methods at 3 months. However, what is important from the point of view of our study: ischemia in the RAH basin occurred after clipping in 15 patients (32.6%), after coiling in only one (2.1%) ! In addition, the authors evaluated frontal lobe ischemia of unconditioned vasospasm in their study and here again ischemic lesions occurred significantly more frequently in clipped patients -⁠ in 12 patients (26.1%), compared to two (4.2%) after coiling. In the Stanford study, only patients with bleeding aneurysms were evaluated. Our incidence of RAH basin ischemia after clipping bleeding aneurysms is 19.6%, 13% less frequently. If we can think of a comparison of our cohort with a recent study by South Korean neurosurgeons published in the prestigious journal Neurosurgery, which evaluated by diffusion-weighted MRI (DWI-MRI) the occurrence of ischemia after surgery of asymptomatic anterior basin aneurysms in 605 patients [15]: in a subset of 108 AComA aneurysms, they observed 18.5% of ischemic lesions overall (16.7% in the whole cohort). The cause of half of them (9.2%) was suspected closure of the perforator(s) (4.6% in the basin of the lateral lenticulostriate perforators from the ACM and 4.6% around the AComA, i.e. incl. Determination of the risk of ischemia for each aneurysm location may be influenced by the fact that a significant proportion of patients (105, or 17.4%) had multiple aneurysms on different vessels, and this was also an independent risk factor for postoperative ischemia. Thus, our five patients out of 83 (6%) with CT-verified RAH ischemia after surgery for asymptomatic AComA aneurysm represent a broadly comparable outcome. However, the mastery of South Korean neurosurgeons is documented by the fact that they distinguish only between the type of cortical or juxtacortical ischemia in the area of surgical dissection (type I), the type of perianeurysmal perforator lesion (type II), which is the ischemia discussed in our paper, the type of distal thromboembolic infarction (type III), and ischemia unrelated to the surgical procedure, localized contralaterally or in another basin (type IV). Obliteration of a large arterial branch, e.g. A2 or M2 clip, was apparently not an option and only 1.8% of ischemias (11/605) were symptomatic in the whole cohort. However, the authors note that perforator (type II) ischaemias are the most common of the four types (48.6%, 53/109) and also the most frequently symptomatic (15.1%, 8/53), with risk factors associated with their occurrence being the application of a temporary clip and the presence of atherosclerotic changes on the vessels surrounding or on the aneurysm sac. For better planning of the procedure and possible avoidance of complications, the precise German authors from the Frankfurt Medical Center (Volker Seifert) suggest imaging of the perforators with 3D rotational angiography before AComA surgery [16].

Conclusion

Our results show that ischemia in the RAH basin was a relatively common complication (14.4%) after AComA aneurysm clipping, especially in the case of bleeding aneurysms, where it reached 19.6%. In the case of asymptomatic aneurysms, ischemia was observed in 6%, but it did not have a lasting negative impact on the clinical condition of the patients. The risk of RAH damage during the operation of bleeding aneurysms is obvious; the reasons for this may be the less clear operative field, the higher risk of intraoperative rupture, and the acute nature of the operation. However, clipping is an essential equipment of the vascular team in complex aneurysms, i.e., knowledge of the spacing, course, position relative to A1 and varieties of RAH is a great advantage for the surgeon. His efforts must be to identify it, preserve its entire course and selectively load the eventual temporary clip on the A1 segment of the ACA without unnecessary temporary occlusion or damage to this perforating artery.

Ethical aspects

The authors declare that the surgery was performed after the patient was informed in accordance with the ethical standards of the Masaryk Hospital Ethics Committee responsible for conducting clinical trials and the Declaration of Helsinki of 1974, revised in 2000. Ethics committee approval or required.

Grant support

This work was supported by the grant of the Agency of Medical Research No. AZV NU22-08-00124.

Conflict of interest

The authors declare that they have no conflict of interest in relation to the subject of the study.

Acknowledgements

The authors would like to thank Monika Němcová for the illustration.