Reperfusion injury from a haemodynamic standpoint and role of pressure-controlled reperfusion in avoiding reperfusion injury

Abstract

Currently, there is no established treatment for post-ischaemic reperfusion-related injury, namely reperfusion injury (RI), which paradoxically exacerbates microvascular and tissue damage in any reperfused ischaemic organ territory. During the ischaemic phase, the autoregulatory apparatus in a subtended organ region temporarily loses its pressure-regulating function because of the combined effects of drastically increased oxygen demand and ischaemic insult. Therefore, it cannot protect the hypoxically injured distal capillary bed from the detrimental effect of the sudden and uncontrolled pressure rise that occurs during the initial phase of reperfusion. This acute capillary barotrauma caused by abruptly initiated reperfusion at systemic pressure can be regarded as an iatrogenic trigger for subsequent damage in the reperfused organ territory. From this haemodynamic perspective, RI can be redefined as a ‘capillary hyperpressurization syndrome’, dictated by the ‘initial reperfusion pressure’. Accordingly, initiating reperfusion gently at lower pressures [pressure-controlled reperfusion (PCR)] and maintaining it at that level until protective autoregulatory myogenic control mechanisms recover may provide substantial benefit in limiting the progressive damage caused by post-ischaemic abrupt and full-pressure reperfusion. In this review, we revisit RI from this haemodynamic perspective and suggest that the same pathomechanism—namely, acute exposure of ischaemically injured microvascular endothelium to an uncontrolled pressure rise during the initial reperfusion phase—predominantly dictates post-reperfusion damage in the heart and in other organ ischaemia–reperfusion settings, where PCR techniques may help limit post-reperfusion damage.

Introduction

Restoration of blood flow to a previously ischaemic organ region may paradoxically aggravate ongoing cellular and microvascular damage. This phenomenon, known as reperfusion injury (RI), is generally considered an unavoidable consequence of post-ischaemic reperfusion, which leads to a further loss of cells that have succeeded to survive after initial ischaemic insult in the subtended perfusion territories, resembling a ‘coup de grâce’. This disruptive cascade begins and progresses rapidly with the sudden post-ischaemic reintroduction of blood flow at systemic pressure to any ischaemic organ territory, which contributes substantially to the magnitude of the final damage.^1,2 Besides the initial ischaemic injury, RI appears to be responsible for as much as the 50% of the final tissue damage in any ischaemia–reperfusion setting.^3,4 Strikingly, the tissue damage incurred during reperfusion may be even greater than the injury due to ischaemia alone such as shown in myocardial ischaemia–reperfusion setting, where ischaemia alone may cause only mild morphological changes⁵ while ischaemia followed by reperfusion leads to a massive microvascular and myocardial injury, oedema and haemorrhage.⁶ RI can occur in all post-ischaemic reperfusion situations such as seen in the treatment of acute myocardial infarction, ischaemic stroke, acute limb and mesenteric ischaemia, chronic thromboembolic pulmonary hypertension, severe carotid or intracranial arterial stenoses, or organ transplantation and may cause additional disability and even death.^7–14 In particular, according to the National Institutes of Health, the global burden of cardiovascular diseases such as myocardial infarction and stroke is increasing rapidly.¹⁵ This trend, in turn, would lead to a continuous rise in the number of patients with RI who would require new and effective treatment options.

Despite this paramount importance and high prevalence, there is neither clear consensus on the underlying pathophysiological mechanisms nor a generally accepted treatment option for RI.¹⁶ In this respect, a deeper understanding of the pathophysiological events taking place in the at-risk area, both during ischaemia and upon reperfusion, is a prerequisite for developing new and effective treatments against RI.¹⁷ To this end, we have revisited the functional and structural consequences of the ischaemia- and reperfusion-induced events as they evolve over time and along consecutive segments of the arterial circulation, which eventually determine the magnitude of the post-ischaemic reperfusion damage (Table 1). Based on this perspective, a new classification scheme, in which different manifestations of RI in various clinical settings were grouped according to reperfusion status, was also proposed in this review (Table 2). This approach mainly demonstrated that, irrespective of whether RI occurs in acute, chronic, controlled or uncontrolled ischaemia–reperfusion settings, its mechanistic nature and consequences share large similarities. This critical insight further suggests a common underlying mechanism behind the different manifestations of RI across various organ systems. As supported by coherent findings from diverse ischaemia–reperfusion scenarios presented below, we propose the joint effect of exhausted pressure-regulating autoregulatory function together with severely disrupted capillary endothelial architecture as the common, predominant pathomechanism underlying RI. Therefore, below we first dissect the fundamental role and position of the autoregulatory apparatus distal to the reperfused artery, as well as the role of the capillary endothelium, both during ischaemia and after reperfusion.

Table 1.

Temporal and spatial changes taking place at the consecutive segments of arterial circulation during severe ischaemia and after reperfusion

	PHASE	Segments				Total vascular resistance	Arterial flow
		Visceral arteries	Arterioles/Autoregulatory apparatus	Capillaries	Cells and interstitium
Time	Early ischaemia	Occluded 2,11,16	Maximally dilated (Adaptive response) 11,18,19	-	Functional abnormality, cellular oedema 20–22	Immeasurably high (Predominantly visceral arterial) 16,23	No-flow 24,25
	Prolonged ischaemia	Occluded 2,11,16	Paralysed (Adaptive response + ischaemic insult) 11,16,26–28	Increased permeability, loss of integrity, hypoxic endothelial injury 21,23,27,29–33	Ischaemic necrosis + cellular oedema 11,20,22,34	Immeasurably high (Predominantly visceral arterial) 11,16,23	No-flow 24,34,35
	Initial phase (early moments) of reperfusion	Re-opened 17,36	Paralysed 11,16,37	Capillary hyperpressurization, disrupted endothelial/tight-junction integrity 27,38,39	Interstitial oedema 40–43	Low (Predominantly microvascular*) 16,26	Over-flow: reactive hyperaemia 16,40
	Late phase of (established) reperfusion	Re-opened 17,36	Partially recovered constrictor response 11,16,28,44	Capillary rhexis 27,45,46	Deepened oedema and haemorrhage causing external capillary compression 41–43,47	Relatively increased (microvascular*) 16,23,41,48	Diminished antegrade flow or “no-reflow” 16,23,24,34,48

Total vascular resistance: Arterial resistance + Microvascular resistance

Table 2.

Principal mechanisms and consequences of post-ischaemic reperfusion damage (reperfusion injury) in different reperfusion settings

	Reperfusion after acute post-occlusive ischaemia	Reperfusion in chronic/severe ischaemia settings	Reperfusion in acute controlled ischaemia settings
Mechanisms
* Hypoxic microvascular endothelial injury, capillary rhexis	(+)6,23,27,33,49	(+)50–53	(+)54–57
* Exhausted arteriolar autoregulation due to both adaptive response and ischaemic insult	(+)11,16,17,37	(+)51,53,58	(+)38,54,59,60
Consequences
Oedema (Cellular/Interstitial)	(+)23,40–43	(+)13,56,61,62	(+)63–66
Haemorrhage	(+)43,46,47,67)	(+)51,52,68,69	(?)38,60,70,71
Clinical settings	Acute myocardial infarction/Primary PCI11,16,17,36,72	Carotid stenosis: staged angioplasty/ stenting51,58,73,74	CABG7,75–77
	Endovascular interventions for acute stroke8,10,20,49	CTEPH: balloon pulmonary angioplasty13,62,78,79	Heart/Lung transplantation9,55,80,81
	Acute peripheral arterial occlusion/ Embolectomy12,71,82,83	Severe intracranial stenosis: stenting84–86	Kidney transplantation87–90

(+) indicates that the mechanism or consequence is documented or present in the referenced literature; (? ) indicates that the evidence is currently limited, inconclusive, or remains a subject of debate in that specific setting.

Autoregulatory apparatus during ischaemic insult and after reperfusion

Arterial networks contain a variety of contractile elements that are critically involved in the myogenic autoregulatory mechanism (autoregulatory apparatus) responsible for regulating microvascular blood flow. There is no consensus regarding the terminology (e.g. pericytes, pre-/peri-arteriolar contractile elements, and other mural cells); therefore, in this review, all of these pressure-regulating elements will be referred to as ‘autoregulatory apparatus’. Autoregulatory apparatus dynamically adjusts parenchymal perfusion pressure to a desirable range as a mutual output of miscellanea of mechanobiological pathways activated in response to changing conditions across organ systems. Results of several studies in various ischaemia-reperfusion settings have reaffirmed that the core pathophysiological mechanism underpinning RI includes the dysfunctional autoregulatory system that cannot protect the already damaged downstream capillaries from the instantaneous pressure overload caused by abrupt reperfusion. In the presence of an acute occlusion or a tight stenosis anywhere in the arterial system, the pressure-regulating myogenic vasoconstrictor function of the autoregulatory apparatus protecting the capillaries from hyperpressurization by preventing transmission of excessive intraluminal pressure across the capillary bed is transiently lost (microvascular stunning) due to a mostly adaptive vasodilator response to deep ischaemia and ischaemic insult^6,91 (Figure 1, Table 1). Critically, it has been shown that myocardial blood flow in the area at risk generally stabilizes around 30 min after reopening an acutely occluded coronary artery following an initial transient rise (post-occlusive reactive hyperaemic flow due to compensatory vasodilation).^24,40 This finding strongly implies that, despite the initial ischaemic insult, the stunned arteriolar smooth muscle cells and, therefore, autoregulatory function can recover within the first 30 min after reperfusion is re-established. Overall, this temporary situation, termed ‘exhausted autoregulation’, leaves the downstream capillary bed totally defenseless against an abrupt pressure rise during the early moments of reperfusion, when blood flow becomes fully pressure dependent. Therefore, when reperfusion is abruptly initiated at systemic pressure after a period of severe ischaemia, ischaemically injured and leaky capillaries, deprived of protection provided by upstream pressure-regulating microvascular elements due to exhausted myogenic control of the autoregulatory apparatus, are inevitably exposed to a sudden pressure burst. This acute capillary hyperpressurization (barotrauma) can be regarded as an iatrogenic trigger for subsequent hyperfiltration oedema and haemorrhage, which appear and progress rapidly shortly after reperfusion.^11,26 Accordingly, a progressive increase in coronary microvascular resistance and zero-flow pressure —which estimates the opening–collapsing pressure of the capillaries—coupled with a corresponding progressive decrease in antegrade coronary flow velocity during the first hour of post-ischaemic reperfusion, as shown previously³⁶ reaffirms the deleterious effect of progressively increasing external compression generated by oedema and haemorrhage on the microcirculation,^23,27,34,92 which, in turn, further impedes antegrade flow to any reperfused territory.^26,36

Figure 1.

Ischaemia and reperfusion induced changes occurring in the affected territory in the settings of abrupt vs. pressure-controlled reperfusion. Schematic summary of the haemodynamic sequence proposed to underlie reperfusion injury. During ischaemia, the autoregulatory apparatus becomes maximally dilated or stunned and the downstream capillary endothelium becomes progressively more permeable and structurally disrupted. Abrupt reperfusion at systemic pressure exposes this unprotected microvascular bed to capillary hyperpressurization, promoting oedema, haemorrhage, rising microvascular resistance, and progressive impairment of antegrade flow. In contrast, pressure-controlled reperfusion applies a lower initial reperfusion pressure, limits early extravasation, allows recovery of upstream pressure-regulating function, and reduces secondary microvascular and tissue injury across organ systems.

Capillary endothelium during ischaemic insult and after reperfusion

Although the impact of ischaemia duration on the magnitude of ischaemia- or reperfusion-related tissue damage varies among different organs,²⁵ in general, duration and severity of ischaemia are the predominant determinants of the overall extent of post-reperfusion damage caused by both ischaemia and reperfusion. Accordingly, as total ischaemic time increases, not only the magnitude of ischaemic injury but also the magnitude of reperfusion-related injury increases. As acute arterial occlusion persists, ischaemia deepens, and severe hypoxia progressively disrupts the microvascular endothelium,⁶ tight junctions,²⁰ and glycocalyx²¹ in the microvascular area at risk, which drastically increases microvascular permeability (microvascular leakage).^6,22 The final stage of this hypoxia-induced capillary disruption may progress to vascular rhexis.⁴⁵ When reperfusion starts abruptly, the already injured and leaky capillaries are inevitably exposed to systemic pressure. Concomitantly, as total ischaemic time increases, the pressure-regulating, vasoconstrictor, and vasoprotective function of the autoregulatory apparatus protecting the downstream capillary bed from hyperpressurization progressively weakens. During the ischaemic period, this combined effect of hypoxia-induced capillary endothelial injury together with exhausted arteriolar autoregulatory function creates the ideal conditions for RI to develop upon reperfusion. In addition, venous drainage of the microcirculation may also be impaired as a result of vascular insult caused by intraluminal microvascular obstruction resulting from microvascular de-novo thrombosis.^{24,35,40,93–97} In this specific milieu, sudden restoration of reperfusion at systemic pressure causing uncontrolled rise in distal intraluminal pressure leads to capillary hyperpressurization and leakage, resulting in interstitial oedema and haemorrhage, which compress capillaries externally and, therefore, contribute to elevated resistance against antegrade flow to the reperfused area, compromising the success of revascularization. In the coronary setting, microembolization—either spontaneous from plaque or thrombus fragmentation or iatrogenic during PCI (guidewire passage, balloon dilation, thrombus aspiration, or stent deployment)—may further aggravate distal microvascular obstruction and myocardial injury, thereby amplifying the reperfusion-injury cascade rather than replacing the central role of capillary hyperpressurization.^37,95,96,98

Because RI, by definition, begins with the initiation of reperfusion, controlling the initial reperfusion period may be the most crucial step in achieving tangible success in avoiding RI. To this end, the potential role of PCR in limiting RI is discussed in detail below.

Rationale of pressure-controlled reperfusion

On this background, the question of ‘how to open’ is just as vital as ‘when to open’ in determining outcome. Since ischaemic injury is dictated by the conditions of the ischaemic phase, it is reasonable to assume that RI is dictated mainly by the conditions of the reperfusion phase. Indeed, most post-reperfusion damage occurs within the first few minutes after reperfusion is established.^28,44 Furthermore, rather than attempting to modulate the cellular and molecular consequences of irreversible tissue damage after abrupt reperfusion at systemic pressure has already begun, modifying the haemodynamic conditions during this initial phase of reperfusion may substantially attenuate RI.

In any acute arterial occlusion scenario, abrupt reperfusion at systemic pressure sharply alters the Starling forces in unguarded capillaries in favour of extravasation of fluid into the interstitium through a weakened endothelial barrier during the early moments of reperfusion, thereby causing hyperfiltration oedema and subsequently haemorrhage, as detailed above. In contrast, restoring blood flow with controlled (lowered) reperfusion pressure during the hyperacute phase of reperfusion may substitute for the temporarily lost myogenic reflex of the arterioles and result in markedly less structural damage, such as oedema and haemorrhage formation. Thus, modifying (lowering) the initial reperfusion pressure via PCR at the time of reopening any acutely occluded or severely stenotic artery supplying a major ischaemic organ territory may help limit the magnitude of capillary damage.⁹⁹

Because the myogenic competence of stunned arteriolar autoregulatory apparatus can recover with time,^24,40 in conjunction with reducing reperfusion pressure during the early moments of reperfusion, allowing sufficient time for paralysed pressure-regulating arterioles to regain their adaptive vasoconstrictor response before restoring full-pressure reperfusion would be a logical approach to prevent an abrupt and uncontrolled pressure rise in the already injured microvascular territory. Importantly, the autoregulatory apparatus should not be considered as a purely myogenic or exclusively arteriolar mechanism, but rather as an integrated microvascular pressure-regulating system encompassing multiple cellular and reflex-mediated components involved in stabilizing capillary perfusion during ischaemia and reperfusion, such as in the brain.¹⁰⁰ In this context, pressure and flow regulation in ischaemic and reperfused myocardium are mediated not only by myogenic mechanisms but also by vasoconstrictor reflex pathways that contribute to coronary resistance modulation during early reperfusion.^11,101,102 Beyond haemodynamic regulation, modulation of reperfusion pressure may also influence metabolic recovery during early reperfusion, which may contribute to the overall tissue response to reperfusion.¹⁰³

Human coronary studies further support the role of reflex sympathetic vasoconstriction during early reperfusion. In patients undergoing coronary stenting, alpha-adrenergic blockade improved coronary flow reserve, myocardial perfusion, and recovery of ventricular function, indicating that reflex vasoconstrictor tone can materially influence microvascular pressure-flow conditions after re-opening of the culprit vessel.^101,102,104 These observations support the haemodynamic concept of PCR by suggesting that post-ischaemic reperfusion pressure is determined not only by conduit opening but also by active neurohumoral modulation within the reperfused territory.

Importantly, clinical reperfusion rarely occurs in pristine vasculature. Ageing, atherosclerosis, hypercholesterolaemia, diabetes, and hypertension are all associated with endothelial dysfunction, vascular stiffening, capillary rarefaction, altered mural-cell or pericyte behaviour, impaired vasomotor reserve, and enhanced alpha-adrenergic constriction.^105–109 In such diseased arterial and microvascular networks, abrupt restoration of flow at full pressure is likely to be even less well tolerated because the pressure-buffering capacity of the microcirculation is already attenuated before the ischaemic insult. Accordingly, the physiological rationale for PCR may be even stronger in real-world patients than in healthy experimental vascular beds, although the optimal target pressure and duration of controlled reperfusion may need adjustment according to the vascular substrate.

Conceivably, the post-ischaemic recovery time for the autoregulatory apparatus could vary between organs as a function of total ischaemic time. Notably, if arteriolar smooth muscle cells cannot regain their adaptive vasoconstrictor response, which could occur with prolonged ischaemic time, the degree of barotrauma caused by abrupt reperfusion on severely disrupted capillaries would be even more disruptive, which may only be mitigated with initial low-pressure reperfusion. In this regard, the extent of RI in any organ system seems to be dictated primarily by the initial reperfusion pressure, which is, in turn, governed by the status and myogenic competence of the autoregulatory apparatus at initial reperfusion phase, when they are indeed in a fully dilated, paralysed position without myogenic competence in response to severe ischaemia. Thus, controlling (lowering) initial reperfusion pressure by PCR appears to be a fundamental step to reduce the final tissue damage.

What pressure would be optimal for pressure-controlled reperfusion?

Under normal circumstances, a highly efficient autoregulatory system delicately recalibrates capillary perfusion pressure within an appropriate range. In the presence of an intact autoregulatory system, the normal capillary perfusion pressure required to maintain organ perfusion is indeed much lower than the low initial perfusion pressure values proposed for the PCR concept.¹⁸ The normal capillary pressure measured at the apex of the capillary loop at heart level ranges from 10.5 to 22.5 mmHg.¹¹⁰ Likewise, the normal value of the zero-flow pressure for coronary arteries, which estimates the opening–collapsing pressure of the capillaries, has been reported to be around 40–50 mmHg at rest in the absence of significantly elevated extramural pressure from oedema or haemorrhage.¹⁹ Consistently, in patients with normal coronary arteries and intact microvasculature, the normal coronary occlusive (wedge) pressure, another parameter that can be used to approximate coronary microvascular haemodynamic status in a post-ischaemic reperfusion scenario, has been reported to be around 22 ± 8 mmHg.^111,112 Naturally, capillary perfusion pressure is much lower in the normal pulmonary circulation (mean pulmonary capillary pressure is around 10 mmHg.¹¹³) From this viewpoint, and consistent with the post-ischaemic reperfusion pressure values proposed in early studies across different ischaemia–reperfusion settings, PCR at 50 mmHg, with a safety margin of 10–20 mmHg, appears to be a rational choice except in the pulmonary circulation, where the optimal re-pressurization value should be further adjusted. Thus, PCR does not actually lower the mean capillary perfusion pressure; rather, it temporarily compensates for the pressure-regulating role of stunned arteriolar resistance vessels while maintaining capillary perfusion pressures comparable to those normally seen in a healthy heart and other organ systems. These observations support the safety of PCR and theoretically invalidate concerns about under-treatment or iatrogenic ischaemia, although optimization of perfusion pressure should be achieved in terms of both safety and efficiency. Additionally, from a practical point of view, an initial reperfusion pressure around 50 mmHg maintained for the first 10–30 min of reperfusion may be theoretically safe and effective in limiting RI in any capillary bed except the pulmonary circulation, which can be used to guide clinical studies designed to implement the technique.

Evidence accumulated over years concordantly suggests that gentle, gradual, or pressure-controlled (low-pressure) reperfusion, rather than a sudden and full-pressure restoration of blood flow in the first few minutes of reperfusion, may radically attenuate post-reperfusion damage in different ischaemia-reperfusion situations (Figures 2–4, Table 3, Supplementary material online, Tables S1–S2, and Supplementary material online, Figure S1). Despite this initial enthusiasm and the uniformly positive preclinical and clinical studies showing the advantages of PCR in post-ischaemic reperfusion situations, technical evolution and clinical adoption have yet to be realized. This is partly due to unresolved questions such as what pressure would be optimal for initial reperfusion, how long reperfusion pressure should be kept below systemic pressure, and whether a fixed-rate or an incremental/ramped algorithm should be followed. In addition, unstandardized methods (flow or pressure control) and technical difficulties in both controlling and maintaining haemodynamic conditions during the early revascularization period seem to be major barriers to further research and development of this promising concept (Figure 5).

Figure 2.

Common regional ischaemia–reperfusion (IR) scenarios in the human body. Illustrative overview of representative clinical ischaemia–reperfusion scenarios across organ systems discussed in the manuscript. Examples include myocardial infarction treated with primary PCI, CABG, and transplantation; lung transplantation and balloon pulmonary angioplasty in chronic thromboembolic pulmonary hypertension; cerebrovascular revascularization such as carotid interventions and thrombectomy; acute limb ischaemia; and renal transplantation. The figure is intended as a conceptual map of where abrupt vs. modulated reperfusion may be relevant rather than an exhaustive epidemiologic classification.

Figure 4.

Pressure-controlled reperfusion and the heart. Synthesis of experimental and clinical evidence supporting pressure-controlled reperfusion in the heart. Across animal models and human studies, gentle or low-pressure reperfusion is associated with better preserved microvascular integrity, less oedema or haemorrhage, smaller infarct size, and better ventricular function than abrupt full-pressure reperfusion. The lower panel indicates the range of supporting evidence from rodents to large animals and humans and illustrates that the proposed target pressure lies below systemic arterial pressure but above the ischaemic threshold.

Table 3.

Controlled reperfusion in heart (selected studies)

Study		Initial reperfusion modification		Meaning
Author (year)	I/R Model	Pressure/Flow	Duration	Main findings
Chandra et al. (1976)114	Pig, Global ischaemia	50 mmHg	N/A	Reduced myocardial oedema and improved recovery after aortic unclamping
Okamoto et al. (1986)115	Dog, LAD ligation	40–50 mmHg	20 min	Better contractile recovery, better preserved microvascular and cellular architecture, less severe myocardial damage, less severe oedema
Yamazaki et al. (1986)59	Dog, LAD occlusion	20 ml/min	120 min	Accelerated improvement in cardiac function and reduced arrhythmias in the early reperfusion period
Sawatari et al. (1991)116	Neonatal lamb	20–40 mmHg	20 min	Preserved endothelium dependent myogenic response
Beyersdorf et al. (1991)117	Human, CABG	25–50 mmHg	20 min	Early recovery of regional wall motion
Ueno et al. (1993)118	Rabbit	Gradual Pressure Increment	N/A	Better systolic function and less interstitial oedema
Pisarenko et al. (1993)119	Guinea pig	N/A	N/A	More effective recovery of aerobic metabolism and better contractility
Sato et al. (1997)38	Dog, LAD ligation	Gradual Pressure Increment	30 min	Smaller infarct size (by 33%), preserved endothelial function
Osaki et al. (2006)120	Pig, transplantation	40 mmHg	20 min	Better functional recovery
Bopassa et al. (2007)121	Rat, Global ischaemia	50 mmHg	5–60 min	Improved postischaemic contractile dysfunction and attenuated necrosis
Musiolik et al. (2010)122	Pig, LAD hypoperfusion(90% attenuated)	Gradual Pressure Increment	30 min	Reduced infarct size by ∼20–36%
Ferrera et al. (2015)75	Rat, Global ischaemia	50 mmHg	10 min	Smaller infarct size and alleviated myocardial necrosis
Sezer et al. (2022)36	Human STEMI– pPCI	70 mmHg	30 min	Smaller infarct size, better preserved microvascular integrity, and blood flow

Figure 5.

Practical questions in pressure-controlled reperfusion application. The left panel summarizes the main variables that can be adjusted during initial reperfusion, including pressure vs. flow control, fixed vs. ramped re-pressurization, target pressure, duration, and expected clinical applicability. The right panel illustrates a catheter-based coronary approach in which a pressure wire is advanced distal to the culprit lesion and partial balloon deflation is used to maintain a controlled distal pressure before full reopening and stent implantation. Distal pressure is monitored continuously during the controlled reperfusion interval.

Below, we scrutinize the potential role of controlled reperfusion strategies that have been implemented to prevent RI mainly in the heart and briefly in other ischaemia–reperfusion settings. Critically, other mechanistic manoeuvres that were tested and found ineffective in preventing RI including ischaemic (pre/post/per) conditioning and deferred stenting in patients undergoing pPCI do not allow a gentle/low-pressure reperfusion but still cause an abrupt pressure rise in the already-injured capillary territory, and hence they should not be mistaken for the proposed mode of pressure-controlled reperfusion (Figure 3).

Figure 3.

Comparison of ischaemic conditioning manoeuvres and deferred stenting with pressure-controlled reperfusion. I: Ischaemia, R: Reperfusion. Conventional intervention, ischaemic post-conditioning, remote pre-conditioning, ischaemic per-conditioning, and deferred stenting all ultimately re-establish coronary flow abruptly at or near systemic pressure in the target vascular territory. By contrast, PCR uniquely modifies the haemodynamic conditions of the first minutes of reperfusion by maintaining a deliberately lower initial reperfusion pressure before full-pressure reperfusion is allowed.

The use of controlled-reperfusion strategies in heart IR setting (Table 3)

Clinical manifestations:

• (a) Primary percutaneous coronary intervention (pPCI), (b) coronary artery by-pass grafting (CABG), and (c) transplantation

Between 1986 and 1991, Buckberg et al. published an extensive set of studies on controlled reperfusion in the heart ischaemia–reperfusion setting, modifying reperfusion mainly by adjusting the flow rate or pressure during the initial reperfusion phase. One of the very early studies,¹¹⁵ in which dogs underwent 4 h of left anterior descending artery ligation followed by reperfusion, showed that abrupt initiation of reperfusion at systemic pressure was disruptive and associated with destruction of microvascular and cellular architecture in the area at risk, more subepicardial oedema, and less potential contractile reserve compared with gentle reperfusion at 40–50 mmHg (Figure 5). Supporting studies conducted in different animal coronary occlusion/reperfusion models including dogs,^38,39,59,115 pigs,^114,120,122 rats,^75,121 guinea pigs,¹¹⁹ rabbits,¹¹⁸ and lambs¹¹⁶ compared controlled post-ischaemic reperfusion techniques using lower initial reperfusion pressures¹¹⁵ or gentle flow⁵⁹ over the first 10–30 min of reperfusion, followed by a gradual or direct increase of the perfusion pressure to the systemic level^38,122 (Table 3). All of these studies consistently demonstrated that a period of controlled reperfusion preceding full-pressure reperfusion protects the microvascular architecture,¹¹⁵ preserves endothelial integrity,^38,116 attenuates oedema and haemorrhage,^{36,114,115,118} lowers the extra-capillary compressive forces,³⁶ and limits the final tissue damage and infarct size^36,38,115 (Figure 4, Table 3). The clinical benefits are manifest as less frequent reperfusion arrhythmias⁵⁹ and improved ventricular contractility.^{36,115,117,121} Additionally, in experimental ischaemia–reperfusion models, PCR has been associated with lower end-diastolic wall thickness¹²³ and pressure,¹²⁴ less calcium overload,^123,125 less ultrastructural damage,¹²⁶ and better myocardial function and recovery in the post-ischaemic myocardium.¹²⁷ In particular, PCR applied immediately at the onset of reperfusion for at least 10 min has been shown to significantly limit infarct size.¹²¹ Moreover, mitochondrial permeability transition pore opening, a key mediator of reperfusion injury, was also shown to be inhibited with PCR.¹²⁸ Taken together, these findings indicate that RI, long thought to be an unavoidable consequence of reperfusion, can indeed be limited by careful control of initial reperfusion pressure.

Another clinically relevant contributor in STEMI and PCI is coronary microembolization, which may occur spontaneously from plaque or thrombus fragmentation or iatrogenically during guidewire passage, balloon dilation, thrombus aspiration, or stent deployment.^37,98 Coronary microembolization can aggravate microvascular obstruction and periprocedural myocardial injury; however, the inconsistent benefit of aspiration and distal protection strategies suggests that it acts as an important aggravating mechanism within a broader reperfusion-injury cascade rather than as the sole determinant of post-reperfusion damage.^{37,98,129–140}

a) Primary PCI for ST elevation myocardial infarction

Patients presenting with ST-elevation myocardial infarction (STEMI) and treated by primary PCI represent the most common form of ischaemia–reperfusion scenario encountered in clinical practice. Hence, both preclinical and clinical research has predominantly focused on addressing the myocardial RI problem. However, almost all major randomized clinical trials in the myocardial ischaemia–reperfusion setting have failed to show any benefit in limiting final infarct size or improving patient outcomes, or have provided conflicting results.^141–163 RI is, indeed, a multifactorial, complex problem that has been attributed to a variety of factors and mechanisms such as oxygen free radicals,^49,164–169 altered calcium handling,^50,170,171 microvascular endothelial dysfunction,^29–33 platelet^35,172–178 and neutrophil plugging,^{32,93,179–186} erythrocyte aggregation,^35,178,187 complement activation,^144,188 atherothrombotic embolization,^41,189,190 in-situ microvascular thrombosis,⁴⁸ and extravascular compression by oedema^42,43 and haemorrhage.^{43,46,47,67,191–200} Major randomized trials addressing the aforementioned potential causes of RI in the myocardial ischaemia–reperfusion setting (primary percutaneous coronary intervention: pPCI), including aspiration thrombectomy,^129–137 distal protection devices,^139,140 mesh-covered stents,^201,202 and deferred stenting,²⁰³ have likewise failed to show benefit in limiting RI or improving patient outcomes. Results of adjunctive pharmacological therapies in this setting have also been controversial and inconsistent, including cyclosporin,^204,205 adenosine-based local adjunctive strategies,^72,206–208 and intracoronary streptokinase.⁷⁶ Moreover, important attempts to reduce myocardial reperfusion injury by mechanical adjunctive strategies^{129–136,201} and pharmacological adjunctive strategies^77,204–206 have failed to show any tangible benefit in limiting myocardial injury or improving clinical outcomes. Given these repeated failures, the pathophysiological mechanisms targeted therapeutically as presumed drivers of RI have come into question again. At least in theory, avoiding or reducing RI-induced damage should only be possible by taking preventive measures just prior to or concurrently with reperfusion. However, to date, potential preventive measures taken even before reperfusion was initiated, such as remote ischaemic conditioning²⁰⁹ and intravenous beta-blocker (metoprolol) administration,²¹⁰ have failed to demonstrate meaningful clinical benefit in patients with STEMI undergoing pPCI. All of these negative results strongly suggest that the main mechanisms triggering the cascade of events leading to RI may indeed be different from those being targeted.

In this regard, despite the well-recognized cardioprotective effects of PCR in experimental myocardial ischaemia–reperfusion models as discussed above, until recently, there had been no clinical trial examining the potential benefit of this concept in patients with STEMI. A recent study by our group examined whether using PCR via the gradual reopening of a totally occluded infarct-related artery could limit microvascular injury in patients undergoing pPCI.³⁶ In this study, patients with a totally occluded infarct-related artery undergoing pPCI were randomly assigned to either standard pPCI with immediate stenting, or PCR with delayed stenting performed at 30 min after initial low-pressure reperfusion (70 mmHg). Continuous assessment of coronary haemodynamics with intracoronary pressure and Doppler flow velocity, including a final measurement of zero flow pressure (Pzf) at the end of the 60-minute monitoring period (primary endpoint) were performed. In the standard pPCI group, progressively deteriorating microvascular perfusion, as evidenced by progressively increased hyperaemic microvascular resistance, resulted in a significantly higher zero flow pressure (Pzf) value at 60 min, indicating greater external compression on the microcirculation. The enzymatic infarct size was also significantly larger in the standard pPCI group compared to the PCR group. Given the paucity of research on effectively limiting reperfusion injury (RI), this study is noteworthy for being one of only two that have attempted to reduce RI-related damage using a catheter-based modulation of initial reperfusion pressure (PCR) during pPCI.^36,70 These findings also reiterate that the mode of post-occlusive reperfusion pressure may be a critical determinant in controlling the post-reperfusion damage.

b) Coronary artery bypass grafting (CABG)

Cardiac surgery with aortic clamping/declamping seems to provide an ideal ischaemia-reperfusion scenario in which the reperfusion period can easily be modified. As early as in 1972, Cooley et al.²¹¹ reported an extreme form of ischaemia–reperfusion damage called ‘stone heart’ representing one of the earliest reports about the detrimental effect of uncontrolled reperfusion. A far less severe but much more common complication of cardiac surgery are the changes in septum function after aortic clamping/declamping. Septal injury due to RI occurs in approximately 40% of patients undergoing CABG and in 60% after valve procedures,²¹² and it has been considered an ‘expected’ outcome/complication of cardiac surgery. In cardiac surgery, minimizing O₂ demand by total heart decompression during regional reperfusion is essential to minimize ischaemic damage;²¹³ however, it is evident that unloading alone may not be enough to prevent the myocardial damage caused by reperfusion initiated suddenly at full pressure. In contrast to the aforementioned various concordantly positive animal coronary occlusion—reperfusion studies, the only randomized trial that compared initial reperfusion with 50 vs. 75 mmHg during the first 2 min following CABG failed to show a significant benefit of PCR in terms of post-operative cardiac output.⁵⁴ Perhaps, the potential benefit of PCR on top of the ventricular unloading and cardioplegia could be too small and/or 2 min of low-pressure reperfusion could have been too short to create a significant difference. Additionally, total heart decompression during surgery minimizes ischaemic microvascular damage, which could have reduced post-reperfusion capillary injury. Thus, CABG procedures, where the left ventricle was already unloaded, may not be the ideal platform to reveal the potential benefit of PCR.

c) Cardiac transplantation

After cardiac transplantation,⁸⁰ an initial coronary reperfusion pressure >50 mmHg has been shown to cause rapidly accumulating oedema and subsequent poor left ventricular contractility^63,214 while the best recovery is observed in non-beating donor hearts resuscitated by continuous myocardial perfusion at 40 mmHg for 20 min. Likewise, in pig hearts arrested by cardioplegia, preservation by continuous myocardial perfusion at 40 cm H₂O provided better functional recovery, improved coronary flow and lower vascular resistance than perfusion at 80 cm H₂O.⁶⁴ These findings consistently indicate that myocardial reperfusion after an acute ischaemic period needs to be performed in a gradual, pressure-controlled fashion in order to prevent an abrupt increase in capillary hydrostatic pressure and to avoid interstitial oedema and intramyocardial haemorrhage formation which are the main components of RI.

Role of pressure-controlled reperfusion in different clinical manifestations of reperfusion injury in other organ systems

Evidence supporting the potential role of controlled reperfusion strategies in mitigating RI-related damage can also be observed especially in the lung, brain and limb ischaemia–reperfusion settings. In this regard, while uncontrolled reperfusion caused severe RI in the lung grafts of rabbits,²¹⁵ rats,⁵⁵ and pigs,²¹⁶ PCR with lower pressures significantly mitigated the lung damage^55,56,61,217 even when it was applied only during the first 10 min of reperfusion.⁸¹ Likewise, in renal transplantation cases, PCR with initial low reperfusion pressure via gradual declamping was shown to be associated with better preserved grafts, significantly improved renal function and reduced postischaemic renal injury.⁸⁷ RI is also a commonly encountered phenomenon following balloon pulmonary angioplasty (BPA) (in 60% to 61% of cases),^78,79 which is the major treatment approach for chronic thromboembolic pulmonary hypertension. Recently, the efficacy of intravascular ultrasound or pressure-wire-guided stepwise, gradual and modulated reperfusion techniques (pressure distal to target lesion < 35 mmHg) was shown during BPA.⁶²

Likewise, cerebral hyperperfusion syndrome, corresponding to cerebral RI, characterized by patchy or diffuse oedema and haemorrhage, may hyperacutely complicate cerebrovascular revascularization processes such as carotid stenting,⁵¹ thrombectomy in acute ischaemic stroke,^10,52,218 and stenting of severe symptomatic intracranial stenoses.^84–86 The most widely recognized mechanism behind cerebral RI, similar to cardiac RI, is the acute increase in cerebral blood flow in the ischaemic microvascular territory, which is not immediately counteracted by the adaptive vasoconstrictor response of fully dilated/paralysed cerebral arterioles.^58,68 This loss of arteriolar protection leaves the subtended ischaemically injured microvascular area totally defenceless against an abrupt pressure rise upon reperfusion.^53,69 Allen and Buckberg¹⁴ provided further evidence that post-reperfusion cerebral protection also seems to be optimal at around 50–60 mmHg. Based on these observations, staged and pressure-controlled angioplasty and stenting have been proposed to allow a time-window for gradual recovery of cerebral autoregulation in tight carotid stenoses.^73,74,219

Furthermore, in acute peripheral arterial occlusion cases, suddenly re-established blood flow at full systemic pressure in deeply ischaemic skeletal muscle, where pressure-regulating contractile elements are fully dilated, exposes already injured capillaries to a pressure burst that may subsequently result in massive oedema.^220,221 In fact, irrespective of the organ regions subjected to RI, the pathomechanism and anatomical pathology of the resulting tissue damage are more or less the same as in limb ischaemia–reperfusion, which is referred to as compartment syndrome. The only two factors that differ are the compliance of the surrounding tissue and its susceptibility to ischaemia, which may vary in different organs. Therefore, in anatomical terms, RI-associated damage can also be viewed as a form of ‘compartment syndrome’ across different manifestations of ischaemia–reperfusion. Accordingly, after acute limb ischaemia, unlike abrupt reperfusion at systemic pressure (100 mmHg for 30 min), PCR (∼50 mmHg for 30 min, then systemic pressure) has been associated with less severe manifestations of RI,¹² including less severe oedema, immediate recovery of contractile function, greater viability, and a significantly smaller increase in leg volume.^65,82 Although the only randomized trial, which enrolled 174 patients with acute limb ischaemia, was not able to show a meaningful difference in amputation-free survival between pressure-controlled reperfusion and conventional treatment groups, PCR techniques have been successfully implemented in patients with severe, prolonged limb ischaemia^83,222,223 (see Supplementary material).

Conclusion

Conventional reperfusion strategies mainly focus on restoring blood flow in the conduit arteries as quickly as possible without taking into account ischaemia-induced particularities in the microcirculation of the subtended tissue. Disregarding this aspect may inadvertently exacerbate RI with deleterious prognostic impact, making it a major therapeutic target in its own right. In this review, we propose that the joint effect of ischaemically damaged capillary architecture and exhausted pressure-regulating myogenic function of the autoregulatory apparatus leaves the already injured distal capillary bed unable to withstand the abrupt hyperpressurization that inevitably occurs at the initial phase of reperfusion, thereby providing a shared mechanism across different manifestations of RI. From this perspective, RI may be redefined as ‘capillary hyperpressurization syndrome’, dictated by the initial reperfusion pressure. Therefore, to achieve tangible success in avoiding RI, capillary hyperpressurization should primarily be avoided by modulating haemodynamic conditions during the initial reperfusion period with PCR techniques rather than by attempting to modulate the sequelae of irreversible post-reperfusion damage that has already been triggered by abrupt, full-pressure reperfusion. Based on this perspective, considering the initial reperfusion phase as a unique treatment window, modulating (lowering) initial reperfusion pressure during the early moments of reperfusion and maintaining it at that level (above the ischaemic range) for a period of time appears to be a promising strategy to limit or avoid the incremental damage caused by RI in any reperfused ischaemic organ territory. Translating this concept into clinical practice will require appropriately designed randomized trials in different ischaemia–reperfusion settings, in which the optimal reperfusion pressure and the optimal duration of low-pressure reperfusion can be tested.

Data availability

This article is based exclusively on previously published data. No new data were generated or analysed; therefore, data sharing is not applicable.

Supplementary material

Supplementary material is available at European Heart Journal Open online.

Author contributions

Murat Sezer (Conceptualization, Investigation, Methodology, Project administration, Resources, Writing—original draft, Writing—review & editing [lead]), Ahmet Tas (Conceptualization, Methodology, Resources, Visualization, Writing—original draft, Writing—review & editing [lead]), Yaren Alan (Conceptualization, Investigation, Resources, Visualization, Writing—original draft, Writing—review & editing [lead]), Ilke Kara Tas (Investigation [lead], Writing—original draft, Writing—review & editing [supporting]), Alp Ozcan (Investigation [supporting], Writing—original draft, Writing—review & editing [lead]), Divaka Perera (Methodology, Writing—original draft, Writing—review & editing [supporting]), Irem Sezer (Methodology, Writing—original draft, Writing—review & editing [supporting]), Sayan Sen (Investigation, Writing—original draft, Writing—review & editing [supporting]), Jan J Piek (Investigation, Writing—review & editing [lead], Writing—original draft [supporting]), and Sabahattin Umman (Conceptualization, Investigation, Supervision, Writing—original draft, Writing—review & editing [lead])

Funding

None declared.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work, the authors used [Bard (Gemini)/Google] solely for the grammatical corrections. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.