**3. Histopathology**

Wang et al. demonstrated that acute coronary occlusions leading to segment elevation myocardial infarction (STEMI) seem to predominately occur in predictable spots within the proximal third of the coronary arteries and that for each 10 mm increase in distance from the ostium, the risk of an acute coronary occlusion significantly decreased by 13–30%, depending on the coronary vessel [17].

In contrast to this, sparse information exists concerning the genesis of CTO and its regional distribution in terms of recanalization. In some publications, soft plaque rupture during acute coronary syndrome (ACS) with rapid thrombotic occlusion followed by its organization is described as the main cause of CTO and only a few appear to derive from atheroma progression [18]. Furthermore, it seems that once thrombotic occlusion occurs the thrombus tends to disseminate retrograde from the site of occlusion to the proximal segments of the vessel with a major side branch [19]. It is known that due to increased chronic hypoxic induction of neovasculature, the affected vessel segment stays biologically active and shows a marked heterogeneity in compensatory angiogenesis with an unpredictable wide range of coronary collateral circulation [20].

#### **3.1. Collaterals and microchannels in CTO**

Successful guidewire crossing may be facilitated by the presence of intravascular microchannels, but structural changes over time with variable localization of these microvessels are not well understood in terms of CTO recanalization [21].

In a post-mortem study of 96 CTO lesions, 49% exhibited residual <99% lumen stenosis by histologic criteria despite angiographically documented total occlusions [22]. In this cohort, adventitia and intimal plaque of total occlusions were the prevalent zones of inflammation and neovascularization. Furthermore, the results revealed in CTOs of all ages a close relation between cellular inflammation and vessel wall neovascularization in terms of location and intensity with an increase in numbers of neovascular channels rather than with an increase in their size.

Munce et al. found two histological types of microvessels in a rabbit model with induced femoral occlusion: a circumferentially oriented "extravascular" and a longitudinally oriented "intravascular" one. Interestingly, extravascular vessels around the occluded artery developed to a maximum at an early time point, followed by a slow regression over time, while intravascular vessel formation within the central body of the occlusion was delayed, and these vessels became thinner and more tortuous over time. Strongly angulated connections between the intra and extravascular microvessels were constantly present, which could explain deviation of the guidewire into extravascular channels during CTO recanalization [21].

Katsuragawa et al. found different histomorphological features in CTOs with tapering of the proximal occlusion point compared to those with a blunt proximal cap [19]. A total of 80% of the tapering-type lesions had shorter occluded segments and showed small recanalized areas with surrounding loose fibrous tissue along the occluded segment. In lesions with a blunt proximal cap, recanalization was rare, and a side branch was frequently found proximal to the occluded segment and easily entered by the guidewire, instead of the occlusion. These features influence penetration of the proximal cap and crossing of the guidewire through the occluded segment and explain why the tapering type of occlusion is favorable for angioplasty.
