**4. Coronary collaterals**

Collaterals are interarterial connections that exist during prenatal development of the coronary circulation and regress in most individuals [26]. They develop in a native occluded vessel through positive remodeling. With the low postocclusive pressure regions being interconnected by collateral vessels, pressure gradient along the occluded segment causes pulsatile shear stress and activates proliferation of vascular smooth muscle and endothelial cells. A complex interplay of actin-binding proteins, integrins and connexions, transcription factors, and mitogen-activated kinases finally leads to an increase in vascular diameter and tissue mass (positive remodeling), but still, the degree of functional restoration of blood flow capacity remains incomplete and ends at approximately 30% of maximal conductance in coronary vessels [27, 28].

The diameter of interarterial connections is usually below the spatial resolution of modern digital angiographic imaging systems (>200 μm) and ranges between 40 and 200 μm. Most of these connecting microvessels have been observed to be located intramyocardially, and only few reach the size of coronary side branches well above 1 mm in diameter [29].

It seems that occluded coronary arteries do not exclusively determine the level of functional collateral flow capacity and that some individuals without stenotic lesions do have immediately recruitable collateral flow to prevent myocardial ischemia during a brief coronary occlusion [30]. However, in patients without well-developed pre-existing interarterial connections, collaterals require between 2 and 12 weeks to fully reach functional capacity [31]. After successful CTO PCI, collateral function usually regresses in collaterals with small diameters but has the potential to recover in case of reocclusion [32].

#### **4.1. Classification of collaterals in CTO**

In 1985, Rentrop et al. developed an angiographic grading system to rate the effect of collaterals in filling the occluded arterial segment [33]. It distinguishes four degrees of collateral recipient artery filling by radiographic contrast medium, but in CTO with welldeveloped, spontaneously visible collaterals, it lacks further differentiation because most collaterals are Rentrop grade 3 (complete epicardial filling by collateral vessel of the target artery). The Werner classification adds an additional parameter to describe spontaneously visible collaterals and demonstrates a close association with clinical determinants of collateral adequacy [31]. Werner et al. graded collateral connections (CC) according to the angiographic visibility: CC0: no continuous connection between donor and recipient artery, CC1: continuous, thread-like connections ( ≥0.4 mm), and CC2: continuous, small side branch-like size of the collateral throughout its course (>1 mm). These CC grades are more practical to determine interventional collaterals suitable for retrograde CTO PCI (**Figure 2**).

**Figure 2.** Interventional collaterals according to the Werner classification.

#### **4.2. Assessment of collateral function**

**4.1. Classification of collaterals in CTO**

**Figure 2.** Interventional collaterals according to the Werner classification.

PCI (**Figure 2**).

48 Interventional Cardiology

In 1985, Rentrop et al. developed an angiographic grading system to rate the effect of collaterals in filling the occluded arterial segment [33]. It distinguishes four degrees of collateral recipient artery filling by radiographic contrast medium, but in CTO with welldeveloped, spontaneously visible collaterals, it lacks further differentiation because most collaterals are Rentrop grade 3 (complete epicardial filling by collateral vessel of the target artery). The Werner classification adds an additional parameter to describe spontaneously visible collaterals and demonstrates a close association with clinical determinants of collateral adequacy [31]. Werner et al. graded collateral connections (CC) according to the angiographic visibility: CC0: no continuous connection between donor and recipient artery, CC1: continuous, thread-like connections ( ≥0.4 mm), and CC2: continuous, small side branch-like size of the collateral throughout its course (>1 mm). These CC grades are more practical to determine interventional collaterals suitable for retrograde CTO Generally, collateral circulation in CTO is predominantly systolic and provides only approximately 50% of antegrade coronary flow, which itself is predominantly diastolic [34]. The assessment of collateral function in CTOs has a different quality than in nonoccluded lesions. Collateral blood pressure distal of a chronic occluded vessel is assessed by placing a piezo-resistive transducer beyond the occlusion, while the antegrade flow has not yet been re-established. This can be ensured by passing occlusive microcatheters over a recanalization guidewire and then exchanged for the pressure wire [29].

#### *4.2.1. Collateral flow index (CFI)*

Intracoronary (IC) flow velocity or pressure measurements to determine collateral flow is theoretically based on the fact that velocity or perfusion pressure signals with values above central venous pressure (CVP) obtained distal to an occluded vessel originates from collaterals [35]. Measurement of such signals provides the variables for the calculation of a CFI, which expresses the amount of flow via collaterals to the vascular region of interest as a fraction of the flow via the normally patent vessel. In contrast to qualitative assessment of collaterals, such as ST-segment changes and chest pain during PCI or the degree of collateral circulation on angiogram prior to PCI, intravascular flow velocity and pressure determination precisely reflect collateral blood flow. Approximately one-third of collateral flow to the occluded area relative to the patent vessel flow is needed to prevent myocardial ischemia at rest [28]. Noteworthy, the majority of patients with MI do not have enough of the collateral flow to avoid ischemia during coronary occlusion [36] and only 10% seem to have a recruitable CFI ≥ 0.4 [36]. Insufficient collateral flow indicated by a CFI ≤ 0.25 independently predicts long-term cardiac mortality [37], and only 10% seems to have a recruitable CFI ≥ 0.4 [36]. Above that, individuals with CTOs tend to have a higher CFI than those without, and the area at risk of myocardial infarction seems to be significantly associated with CFI.

#### *4.2.2. Fractional flow reserve in the donor artery and coronary steal*

Microvascular vasodilation might lead to reduced collateral blood flow during physical or pharmacological provocation in individuals with collateral-dependent blood supply. In order to generate coronary steal, Werner et al. describes, in reference to Gould et al., the following assumption: epicardial stenosis of the donor artery causes a pressure drop proximal to the collateral origin; the collateral resistance is significant, and the microvasculature distal to the occlusion lacks a vasodilatory reserve due to being already maximally dilated [38].

Therefore, Werner et al. measured fractional flow reserve in the donor artery (FFRD) at the origin of the collaterals in patients with CTO and recorded coronary flow velocity and pressure during recanalization. Patients with steal had more severe regional dysfunction and those with steal but without an FFRD < 0.8 tended to have an impaired microvascular function. The authors concluded that coronary steal mainly occurs as a result of hemodynamically significant donor artery lesions and might have an adverse effect on the preservation of myocardial function by collaterals.

In 50 patients who successfully underwent CTO recanalization compared to 50 matched non-CTO PCI subjects, patients with CTO and an intermediate donor artery stenosis showed a low FFRD with a high frequency of ischemia in the donor artery territory, which was often normalized by successful CTO treatment, thus suggesting recanalization of CTO as a preferred therapeutic strategy. Reference: CCI 2014.
