Calcified Coronary Artery Disease: Pathology, Prevalence, Predictors and Impact on Outcomes

Abstract

Calcified coronary artery disease is a common clinical finding and is visible angiographically in 25–30% of patients presenting for percutaneous coronary intervention. The presence of coronary calcium, even without coronary artery obstruction, confers an adverse clinical prognosis. Coronary calcium score on CT is additive in predicting risk of cardiovascular events beyond traditional scoring systems. Deposition of calcium in coronary arteries is initiated by the formation of an atherosclerotic plaque. Thereafter, multiple processes and pathways are involved, resulting in initial microcalcifications that coalesce into calcium sheets. Calcified nodules are thought to occur from rupture of these sheets. Calcified coronary stenoses requiring revascularisation result in greater target lesion failure and overall major adverse cardiovascular events than noncalcified lesions, regardless of the mode of revascularisation. Modifying calcium prior to stenting to optimise stent expansion is required and intracoronary imaging can greatly facilitate not only the detection of coronary calcium, but also the confirmation of adequate modification and stent optimisation. In this review, the authors examine the pathophysiology, prevalence, predictors and impact on outcomes of coronary calcium.

Coronary artery calcification is common and confers adverse cardiovascular outcomes.1,2 Its frequency increases with age and its presence is associated with a higher atherosclerotic plaque burden.2 It can be detected even in asymptomatic patients on CT and higher Agatston calcium scores are closely associated with adverse outcomes.3,4 Its natural history is that of progression from microcalcification in the vessel intima to eventual concentric sheets of circumferential calcium, with rupture of these sheets creating calcified nodules. Microcalcifications are thought to be associated with unstable coronary artery disease (CAD) with higher percentage calcified plaque resulting in a relatively more stable plaque.5 Nodular calcification – particularly eruptive nodules – despite representing a very severe form of calcification, can be an unstable morphological subtype and responsible for acute coronary syndromes.6

Percutaneous intervention in calcified atherosclerotic disease is fraught with difficulty. The non-compliant nature of calcified lesions results in difficulties with plaque preparation that may result in stent under-expansion and, ultimately, stent failure. Furthermore, complications associated with percutaneous coronary intervention (PCI) in calcified CAD are well documented, including flow-limiting dissection and perforation. PCI in calcified lesions is nonetheless becoming more common and up to 25% of patients presenting for planned PCI have angiographically detectable calcium. A number of reasons for this increasing prevalence have been identified, including advanced age because of longer life expectancy and increased risk factors for calcified CAD, such as hypertension, diabetes and chronic kidney disease.7–9 As interventional cardiologists, performing PCI in these patients is unavoidable and an understanding of the underlying pathophysiology and impact on outcomes is therefore essential. This review will focus on the pathophysiology, prevalence and impact on outcomes of coronary calcification and propose a pragmatic calcium modification algorithm guided by intracoronary imaging.

Pathophysiology

Vascular calcification involves the crystallisation of calcium and phosphate which is deposited in the extracellular matrix as hydroxyapatite.10 It is classified based on the location of accumulated calcium deposits, which can be either intimal or medial calcification. Intimal calcification is most commonly seen in coronary arteries where injury to the intima and initiation of a subsequent inflammatory process contribute to atherosclerotic plaque development, while medial calcification is more frequently seen in peripheral arteries and is common in those with renal disease and diabetes. For the purposes of this review, we will concentrate on intimal calcification.

Coronary intimal calcification is a complex and – as yet – incompletely understood process. It is progressive and begins with pathological intimal thickening, progressing to microcalcifications, fragmented calcium and eventually calcium sheets (Figure 1).10 The presence of coronary atherosclerotic plaque is required for the initiation of the calcification process and within this atherosclerotic plaque, smooth muscle cells, macrophages and extracellular matrix vesicles are all thought to play a role. The atherosclerotic process begins with damage to the intima from exposure to LDL-cholesterol, which initiates an inflammatory process.11 Oxidative changes to LDL in the intima result in the release of pro-inflammatory cytokines that attract monocytes, which mature into macrophages and eventually foam cells.11,12 Multiple mechanisms are thought to initiate plaque progression from a lipid-rich to a calcified plaque. Among these mechanisms are the recruitment of vascular smooth muscle cells that undergo osteogenic differentiation promoted by oxidative stress and bone morphogenetic proteins (BMP), namely BMP-2 and BMP-4, resulting in calcium deposition.13 Other mechanisms include smooth cell apoptosis resulting in the formation of calcifying matrix vesicles by macrophages that, if not quickly cleared, can act as a substrate for calcium deposition.14 Subsequent macrophage apoptosis can, in turn, create a favourable milieu for the progression of calcification.5,10,12 Functioning macrophages and vascular smooth muscle cells attempt to inhibit the deposition of calcium by producing matrix Gla protein and fetuin-A that facilitate the rapid phagocytosis of apoptotic macrophages and smooth muscle cells.12 However, eventually an imbalance in calcium-promoting and -inhibiting factors result in calcium deposition.

Figure 1: Progression of Coronary Calcification as Seen on Histology.

Standard haematoxylin and eosin stain demonstrating varying levels of calcium deposition in coronary arteries at low power (A–D) and high power (E–H). A and E: Arrows demonstrate early micro-calcification deposition; B and F: Arrows show established calcium deposition in the form of calcium sheets; C and G: The microcalcifications can be seen coalescing into an established nodule; D and H: There is a spectrum of changes from microcalcifications, sheets and established calcium nodules.

Calcification begins with the deposition of microcalcifications of <0.5 μm in size that progress to punctate calcification of 15 μm to 1 mm. These areas of punctate calcifications coalesce to form fragmented calcium (>1 mm in size), which further coalesce to form sheets of calcium.10 Calcified sheets can extend across several quadrants of the vessel. Extensive calcification results in altered vessel compliance, which contributes significantly to the difficulty in percutaneously treating calcified CAD.15 Nodular calcification is thought to form when rupture of the calcium sheet occurs, causing protrusion into the vessel lumen.

The progression from punctate calcification to sheets of calcium also represents a progression from plaque instability to plaque stability. Smallvolume spotty or punctate calcification is associated with a phase of plaque instability and plaque rupture, while more extensive calcification is associated with stabilisation of the atherosclerotic plaque. Microcalcification with deposits as small as <15 μm has been found to be associated with plaque rupture.

Using postmortem coronary artery samples, Vengrenyuk et al. demonstrated increased local stress, concentrated around minute calcium deposits (~10 μm) in the atheromatous fibrous cap, which were up to twice that of larger calcium deposits.16 This increased stress is thought to contribute to an increased risk of rupture at these sites.16 However, microcalcification deposits are beyond the resolution of current intracoronary imaging modalities and even that of widely available CT scanners. Intracoronary imaging can, however, detect spotty calcification and Ehara et al. first described the association between spotty calcification and plaque instability.17 In this study of 178 patients, intravascular ultrasound (IVUS) was used to detect coronary calcification in those presenting with acute MI, unstable angina and stable angina. Spotty calcium was defined as a calcium arc of <90° and was significantly more common in those presenting with acute MI than other clinical presentations that tended to have more extensive calcium deposits.17 Additionally, those with spotty calcium had associated fibrofatty plaque and positive remodelling.17 Pu et al. examined 2,294 vessel segments in 151 vessels from 62 postmortem patients using IVUS, near-infrared spectroscopy IVUS (NIRS-IVUS) and histology.18 Similar to the findings of Ehara et al., spotty calcification was associated with greater lipid core plaque (fibroatheroma) on NIRS-IVUS compared with lesions with extensive calcification.18

CT studies assessing plaque composition have been consistent with the aforementioned studies using intracoronary imaging. In the PARADIGM registry, serial CT scans in over 2,000 lesions demonstrated a positive association between calcified plaque volume and cardiovascular events, while percentage calcified plaque volume was inversely related to cardiovascular events.19 Both overall major adverse cardiovascular events (MACE) and revascularisation were lower in patients with higher, compared with lower, percentage calcified plaque volume (MACE 9.4% versus 14.6%; p=0.022 and revascularisation 8.8% versus 14.3%; p=0.016).19 Similarly, a sub-analysis of the MESA study, again using CT assessment of coronary artery calcium, demonstrated an association between higher calcium density and lower risk of cardiovascular events.20 As such, as the proportion and density of calcium increases within the plaque, the plaque is stabilised, thereby reducing acute events.

However, calcified nodules may be the exception. Despite having a high calcific burden, they have been associated with acute coronary syndromes. Morphologically, calcified nodules can be described as eruptive or non-eruptive. They are commonly found in areas of high torsion or mobility within the vessel and, as such, are common in the proximal right coronary and circumflex arteries.6,21–23 Non-eruptive nodules have a thick overlying fibrous plaque, while eruptive nodules are those with a disrupted fibrous cap, often with overlying thrombus.6,24 These morphological subtypes often have different clinical presentations and consequences. Eruptive nodules, with disruption of the overlying fibrous layer, have been associated with acute coronary syndromes. In a study by Torii et al., histological examination of 26 nodules from 25 subjects who experienced sudden cardiac death found evidence of disruption to the overlying fibrous cap, endothelial loss and surface thrombus consistent with the eruptive morphological subtype of calcified nodules.6 Similarly, in a sub-analysis of the CLIMA study, a prospective multicentre study including patients undergoing optical coherence tomography (OCT) assessment of the proximal left anterior descending artery during indicated coronary angiography (n=1,003), calcified nodules were found in 17.9% of which 3% were nodules with disruption of the overlying fibrous cap (eruptive).25 At 1 year, the composite endpoint of cardiac death and target vessel MI had occurred in 20% of those with eruptive nodules versus only 2.7% of those nodules with an intact cap.25

Prevalence and Predictors

The prevalence of coronary artery calcification is dependent on the population being examined (symptomatic versus asymptomatic), and the imaging modality being used for assessment (CT, invasive coronary angiography or intracoronary imaging). CT is highly sensitive for the detection of coronary calcium while angiography is insensitive but highly specific. Intracoronary imaging modalities (IVUS and OCT) are more sensitive than angiography.26 A number of risk factors have been identified for the development of coronary calcification (Figure 2). MESA was one of the first large-scale observational studies to examine predictors of coronary calcium progression at a population level.2 Asymptomatic patients from the four major ethnic groups across six US cities (white, black, Hispanic and Chinese) underwent serial CT scans with calcium scoring at 2-year intervals. The incidence of newly detected calcium in this cohort of over 5,000 patients was 6.6% per year. Age was a significant contributing factor with coronary calcium detectable in only 5% of those aged <50 years compared with 12% of those aged >80 years.2 Males more frequently had coronary calcification compared with females, which is likely because of the protective effect of oestrogen in women.2,27 Considerable ethnic variation was also noted, with the white ethnic group having a higher prevalence of coronary calcium. Compared with the white ethnic group, the RRs for having coronary calcification were 0.78 (95% CI [0.74–0.82]) in black, 0.85 (95% CI [0.79–0.91]) in Hispanic and 0.92 (95% CI [0.85–0.99]) in Chinese ethnic groups.2 Hypertension and diabetes were also associated with prevalence and progression of coronary calcification. Other similar population-based studies, including the Framingham Heart Study, the Heinz Nixdorf Recall Study and the Rotterdam study in asymptomatic patients demonstrated similar findings and, at longer-term follow-up, an association between higher coronary calcium scores and cardiovascular events was seen.1,28–30 This is unsurprising given that coronary calcification correlates well with overall plaque burden.19

Figure 2: Risk Factors for Calcified Coronary Artery Disease.

Patients with diabetes have been found to have increased risk of coronary artery calcification in numerous studies. Furthermore, patients with diabetes have increased mortality for any given calcium score on CT compared with those without diabetes.31 The degree of diabetes control is linked to the extent of coronary calcification as measured by glycosylated haemoglobin levels (HbA1c). In the CARDIA study, more than 2,000 patients underwent serial coronary artery calcium scoring over a 5-year period. Higher HbA1c levels were associated with incident coronary calcification (RR 1.45; 95% CI [1.02–2.06]), progression of coronary artery calcification (RR 1.51; 95% CI [1.16–1.96]), and advanced coronary calcification progression (RR 2.42; 95% CI [1.47–3.99]).32

Patients with chronic kidney disease (CKD) are known to have an increased risk of calcified CAD. Initial autopsy studies demonstrated a relationship between lower renal function and the severity of coronary calcification.33 In vivo studies in larger populations have confirmed these findings. Bundy et al. examined over 1,100 patients in the CRIC cohort comprising patients aged 21–74 years with mild-to-moderate CKD and without known CAD to determine risk factors for progression of calcified CAD in this group.34 Subjects underwent serial CT coronary artery calcium scoring and 60% were found to have calcified CAD at baseline. While these patients did have a greater burden of traditional cardiovascular risk factors, a lower estimated glomerular filtration rate, independent of other cardiovascular risk factors, was associated with coronary calcification and its progression.34 As diabetes and CKD often co-exist, it is, therefore, unsurprising that the presence of CKD in patients with diabetes further increases the risk of coronary calcification.35

Those with end-stage renal disease requiring dialysis have increased risk and more rapid progression of coronary calcification than those not requiring dialysis.36 Initial autopsy studies have confirmed increased severity of coronary calcification in patients on haemodialysis, with these findings being confirmed by CT assessment.37,38 A number of factors likely contribute to these findings. Bone-mineral disorder prominent in dialysis patients results in dysregulated calcium and phosphate metabolism, consequently affecting vitamin D and parathyroid hormone regulation. Subsequent use of calcium-based phosphate binders and vitamin D supplementation contribute to increased vascular calcification (both intimal and medial) in these patients.39,40 Strict control of phosphate levels can slow the progression of coronary calcification in dialysis patients, as can the use of low calcium dialysate.41,42

Statin therapy is known not only to reduce atheroma volume and the risk of MACE but also to increase coronary calcification, which may contribute to plaque stabilisation.19,43,44 A serial IVUS study performed by Hiro et al. in patients prescribed statins following an acute coronary syndrome presentation found significant reduction in plaque volume with the use of statins.44 Nicholls et al. confirmed similar findings in a randomised controlled trial comparing the effect by IVUS on plaque volume of high-dose atorvastatin and rosuvastatin.45 Puri et al. performed a post hoc analysis of eight prospective randomised controlled trials evaluating serial changes in plaque composition by IVUS in patients on high-intensity, low-intensity and no statin therapy.43 Those on high-intensity statin therapy had a reduction in percent atheroma volume with a concomitant increase in calcification.43 Similar findings have been confirmed on CT.

In a sub-analysis of the PARADIGM registry, statin therapy at baseline was associated with greater increase in the percentage calcified plaque volume between serial CT scans (mean increase 3.5% ± 7.7% per year). Similarly, commencing statin therapy between scans was also associated with increased percentage calcified plaque volume (mean increase 4.3% ± 6.9% per year).19

The association between exogenous calcium supplementation and coronary calcification has been difficult to determine because much of the data have been obtained from observational studies with many potential confounding factors.46 An analysis of 1,914 patients from the ARIC study (an observational study that enrolled patients aged 45–64 years between 1987 and 1989 with prospective follow-up since that time) found no association between dietary calcium and subsequent vascular calcification.47 Similarly, an analysis of the MESA cohort found no association between dietary calcium intake, but an increased risk of coronary calcium associated with exogenous calcium supplementation.48 Randomised controlled trials specifically addressing the issue of calcium supplementation and coronary calcium are lacking.

Predictors for the presence of coronary calcification in symptomatic patients are similar to those in asymptomatic patients. In a pooled analysis of the HORIZONS-AMI and ACUITY trials, Généreux et al. identified age (per 10-year increase HR 1.29; 95% CI [1.23–1.35]), male sex (HR 1.29; 95% CI [1.12–1.50]), hypertension (HR 1.12; 95% CI [1.02–1.31]) and ST elevation MI presentation (HR 1.55; 95% CI [1.37–1.76]) as independent predictors of the presence of a calcified target lesion by angiographic assessment.7 Similarly, in a study by Huisman et al. (a pooled analysis of the TWENTE and DUTCH PEERS studies), older age and diabetes were more common in those with severe calcification on angiography.9 Pooled analysis of the BIOFLOW studies again demonstrated angiographic moderate-to-severe calcification more commonly in older age.8

Impact on Outcomes Following Percutaneous Coronary Intervention

Calcified CAD is known to confer an adverse impact on prognosis following revascularisation (Table 1). In fact, incomplete revascularisation in these lesion subsets has been noted in many PCI studies, demonstrating a possible reluctance to undertake revascularisation in this complex lesion subset.49 In patients presenting for PCI for various reasons (stable and unstable presentations), the prevalence of significant coronary calcification by angiography is in the region of 25–30%.7–9 Using intracoronary imaging, the prevalence of coronary calcification is significantly higher. Wang et al., in a retrospective single-centre study of patients undergoing PCI with concomitant angiography, IVUS and OCT assessment, demonstrated an almost doubling of the proportion of lesions found to have coronary calcium by intracoronary imaging compared with angiography.26

Table 1: Major Randomised Controlled Trials with Sub-group Analysis of Outcomes in Calcified Lesions.

Study	Study Design	Study Population	Patients Included in the Analysis	Imaging Modality for Calcium Assessment	Outcome
Généreux et al. 20147	Pooled analysis of the HORIZONS-AMI and ACUITY randomised controlled trials	Acute coronary syndrome	6,855: 2,190 severe/moderate calcium 4,665 none/mild calcium	Angiographic assessment	MACE at 1 year: 12.9% none/mild calcium; 16.1% moderate/severe calcium; p=0.001 Death at 1 year: 2.8% none/mild versus 4.6% moderate/severe; p=0.0002 Cardiac death at 1 year: 1.8% none/mild versus 3.1% moderate/severe; p=0.001 Ischaemic TLR at 1 year: 6.0% none/mild versus 8.2% moderate/severe; p=0.002 Ischaemic TVR at 1 year: 7.5% none/mild versus 9.4% moderate/severe; p=0.02 Definite stent thrombosis at 1 year: 1.7% none/mild versus 2.7% moderate/severe; p=0.008
Huisman et al. 20169	Pooled analysis of the TWENTE and DUTCH PEERS (TWENTE II) randomised controlled trials	Stable angina	1,423: 342 severe calcification 1,081 none/mild calcification	Angiographic assessment	MACE‡ at 2 years: 10.2% no severe calcification versus 16.7% severe calcification; p=0.001 POCE† at 2 years: 14.4% no severe calcification versus 22.2% severe calcification; p=0.001 TLF at 2 years: cardiac death, TV-MI, CI-TLR: 9.8% no severe calcification versus 16.4% severe calcification; p=0.001 Cardiac death at 2 years: 1.5% no severe calcification versus 4.4% severe calcification; p=0.03 Target vessel-related MI: 3.4% no severe calcification versus and 7.6% severe calcification; p=0.001 Clinically indicated TVR: 5.8% no severe calcification versus 7.6% severe calcification; p=0.24 Definite stent thrombosis: 0.4% no severe calcification versus 1.8% severe calcification; p=0.02
Hemetsberger et al. 20218	Pooled analysis of BIOFLOW II, IV and V	All-comers population of stable angina, silent ischaemia and acute coronary syndrome	2,361: 303 moderate/severe calcification 1,642 none/mild calcification	Angiographic assessment	TLF (cardiac death, TV-MI, ID-TLR, definite/probable stent thrombosis) at 2 years: 8.4% none/mild calcification versus 13.5% moderate/severe calcification; p=0.003 TV-MI at 2 years: 4.8% none/mild calcification versus 11.2% moderate/severe calcification; p<0.0001 ID-TLR at 2 years: 3.9% none/mild calcification versus 5.0% moderate/severe calcification; p=0.304
Kawashima et al. 202250	Sub-analysis of the 10-year outcomes of the SYNTAX trial	Multivessel disease (stable/unstable angina)	PCI group: 627 without HCL 276 with HCL lesion CABG group: 641 without HCL 256 with HCL lesion
Angiographic assessment	All-cause mortality at 10 years in PCI group: 26% in non-HCL versus 34% in HCL; p<0.001 All-cause mortality at 10 years in CABG group: 8.8% in non-HCL versus 39% in HCL; p<0.001
Sanz-Sanchez et al. 202349	Subgroup analysis of the MATRIX trial||	Acute coronary syndrome (STEMI and non-STEMI)	7,446: 875 severe calcification 6,571 no severe calcification	Angiographic assessment	MACE§ at 1 year: 15% no severe calcification versus 27.1% severe calcification; adjusted HR 1.23; 95% CI [1.06–1.44]; p=0.008 Composite endpoint of CV death, MI, stroke TVR or ST: 4.2% no severe calcification versus 26.4% severe calcification; adjusted HR 1.31; 95% CI [1.11–1.53]; p=0.001

*MACE: death, reinfarction and unplanned revascularisation. †POCE: all-cause mortality, any MI, repeat (target or non-target) revascularisation. ‡MACE: all-cause death, any MI, emergent CABG, CI-TLR. §MACE: all-cause mortality, MI or stroke. ||Revascularisation was both using PCI and CABG. CABG = coronary artery bypass graft; CI-TLR = clinically indicated target lesion revascularisation; CV = cardiovascular; HCL = heavily calcified lesion; ID-TLR = ischaemia driven target lesion revascularisation; MACE = major adverse cardiovascular events; non-STEMI = non-ST elevation MI; PCI = percutaneous coronary intervention; POCE = patient-orientated cardiovascular event; ST = stent thrombosis; STEMI = ST elevation MI; TLF = target lesion failure; TLR = target lesion revascularisation; TVF = target vessel failure; TV-MI = target vessel MI; TVR = target vessel revascularisation.

Interestingly, patients with angiographically moderate or severely calcified lesions have an adverse prognosis regardless of the type of revascularisation undertaken (PCI or coronary artery bypass graft [CABG]).7,8,49,50 CABG in these patients has been traditionally thought to circumvent the problems associated with performing PCI in calcified lesions. However, the recent 10-year analysis of the SYNTAX study demonstrated that CABG in patients with heavily calcified lesions by angiography had a higher all-cause mortality than those without calcified lesions (39% versus 18%; p<0.001).50 This observation is interesting, given that many patients are referred for CABG when coronary calcification is predicted to result in inadequate results or incomplete revascularisation with PCI. The likely explanation of these findings is that severe calcification is essentially a marker of disease progression and is strongly associated with total plaque burden and – unsurprisingly – poorer outcomes.19 As such, the presence of coronary calcification in and of itself is a marker of advanced disease and adverse outcomes are, in part, the result of the unabated progression of the atherosclerotic process and reflect the patient’s co-morbidities such as diabetes and CKD that predispose to coronary calcification.

PCI in calcified lesions is fraught with difficulty and multiple studies have demonstrated poorer results, with target lesion revascularisation (TLR) rates of between 6–8% at 1-year and up to 12–14% at 2 years (Table 1).7,8,49 In the aforementioned SYNTAX study, 10-year mortality among those undergoing PCI with a highly calcified lesion by angiography was 34% compared with 26% in those with non-calcified lesions (p<0.001).50 Similarly, a pooled analysis of the HORIZONS-AMI and ACUITY trials demonstrated increased all-cause mortality (6.3% versus 2.8%), cardiovascular mortality (4.0% versus 1.8%) target lesion failure (TLF; 8.7% versus 6.0%) and overall MACE (19.9% versus 12.9%) at 1-year in those with severe calcification compared with mild calcification on angiography (p<0.01 for all comparisons).7 More recently, a sub-analysis of the MATRIX trial reported an almost doubling of TLR rates in severely calcified versus non-calcified lesions (6.8% versus 3.5%, unadjusted HR 1.99; 95% CI [1.48–2.65]; p<0.001).49 While both of these studies were performed in acute coronary syndrome populations, results are similar in those presenting with stable angina. Huisman et al., in a study of patients with stable angina undergoing PCI, demonstrated higher rates of target vessel failure (TVF; 16.4% versus 9.8%; p=0.001), cardiac death (4.4% versus 1.5%; p=0.03), target-vessel MI (TV-MI; 7.6% versus 3.4%; p=0.001), and definite stent thrombosis (1.8% versus 0.4%; p=0.02) at 2 years in severely calcified versus non-calcified lesions as assessed by angiography.9

Innovation in stent design has made some inroads into improving patient outcomes with second-generation drug-eluting stents (DES) demonstrating better outcomes compared with first-generation DES in the treatment of calcified lesions.51 Among second-generation devices themselves, with different antiproliferative drug and polymer technology, differences are negligible. A pooled analysis of the BIOFLOW studies by Hemetsberger et al. reported overall TLF and TLR rates of 13.5% and 5.0%, respectively, at 2 years for angiographically moderately-to-severely calcified lesions treated with newer generation DES, but no differences in TLF rates between the stenting platforms investigated (bioresorbable-polymer sirolimus-eluting stent and durable-polymer everolimus-eluting stent).8 Drug-coated balloon (DCB) therapy for de novo lesions is currently gaining interest with the added benefit of the ‘leave nothing behind’ strategy circumventing any possibility of stent under-expansion due to calcium. Until now, DCB therapy for de novo lesions has been largely reserved for small vessel disease with comparable results to DES in this cohort.52 However, the use of DCBs in calcified CAD results in greater TLF than noncalcified disease. A recent retrospective study of 328 lesions treated with DCB under OCT guidance found the presence of calcium to be a predictor of DCB treatment failure (risk of TLF per 90° calcium arc on OCT, HR 1.34; 95% CI [1.05–1.72]; p=0.02).53 Further prospective studies are required to evaluate this treatment modality in calcified lesions.

There are several identifiable causes for these documented adverse outcomes in patients with calcified CAD undergoing PCI. Firstly, in a similar fashion to those who undergo CABG, severe calcification is a marker of advanced disease and continued progression of atherosclerosis in coronary vessels and other vascular territories may contribute to events including all-cause and cardiovascular mortality. As mentioned previously, the presence of calcified plaque signifies a higher plaque burden overall and is a marker of disease progression.19 This is evidenced in clinical practice in the study by Généreux et al., where patients with angiographically moderate-to-severe calcified lesions had more extensive CAD as a whole, while target lesions in this group were longer and more commonly complex with higher rates of total occlusions and bifurcations.7

Notwithstanding the overall degree of disease burden in these patients, PCI in calcified lesions may be challenging for a number of other technical reasons that may also contribute to the adverse outcomes. Mechanical damage to the stent during delivery to the target site, damage to the polymer and alterations in drug-elution kinetics have all been identified as potential mechanisms of suboptimal results following PCI.7,54 These potential adverse consequences of stenting may be mitigated by undertaking plaque modification. However, lesion preparation in some instances may result in procedural complications, such as flow-limiting dissection or even perforation.7,55 Furthermore, complications in these patients extend beyond the procedure, with an increased bleeding risk at 30 days in patients undergoing PCI for calcified lesions.56 The risk of periprocedural complications during PCI must be balanced with the risk of stent failure beyond the procedure. Inadequate plaque modification can lead to stent under-expansion and a small final stent area, which are the main predictors of stent failure.57,58

Improving Outcomes in Patients with Calcified Coronary Artery Disease Undergoing Percutaneous Coronary Intervention

Calcium modification guided by intracoronary imaging holds most hope in optimising results and outcomes when performing PCI in calcified lesions. Failure to modify calcium may, in part, be because of a failure to recognise its presence, particularly if PCI procedures are guided based on angiography alone. Correlation between core lab and operator detection of calcium by angiography is known to be weak,8 and angiography is poorly sensitive compared with intracoronary imaging (IVUS or OCT).26,59 Recent developments in intracoronary imaging software have enabled the use of artificial intelligence for the detection and measurement (arc, depth) of calcium. Furthermore, intracoronary imaging is known to improve outcomes by reducing TLR from all-comer populations to more complex lesion subsets including calcified lesions.60–67

More recently, a network meta-analysis of intracoronary imaging (both IVUS and OCT) in >15,000 patients demonstrated not just a reduction in stent failure events, such as TLF (RR 0.71; 95% CI [0.63–0.80]; p<0.0001), TV-MI (RR 0.82; 95% CI [0.68–0.98]; p=0.030), TLR (RR 0.72; 95% CI [0.60–0.86]; p=0.0002) and stent thrombosis (RR 0.52; 95% CI [0.34–0.81]; p=0.0036), but also in all-cause mortality (RR 0.75; 95% CI [0.60–0.93]; p=0.0091) and cardiac mortality (RR 0.55; 95% CI [0.41–0.75]; p=0.0001).68 Although current European and American clinical practice guidelines do not strongly endorse the use of intracoronary imaging specifically for calcified lesions, recent expert consensus documents produced by the European Association of Percutaneous Coronary Intervention and the Society for Cardiovascular Angiography and Interventions on the management of coronary calcium recommend the use of intracoronary imaging for the assessment of plaque morphology and optimisation of PCI.69,70

Intracoronary imaging guidance for PCI in calcified vessels should be employed at all stages, including baseline assessment, after plaque modification and following stenting, to optimise the result.71 The optimal choice of intracoronary imaging modality remains debated. A study by Wang et al. suggested that IVUS may be marginally more sensitive for the detection of coronary calcium than OCT.26 In this study, 440 lesions were assessed by angiography, OCT and IVUS. While both OCT and IVUS were more sensitive than angiography for the detection of coronary calcium, 13.2% of calcium detected by IVUS was either not detected or underestimated by OCT. Overlying tissue limiting penetration of OCT was the predominant reason for the lack of detection or underestimation of calcium.26 IVUS, too, is susceptible to this phenomenon. In a postmortem study by Pu et al., IVUS failed to detect histologically evident calcium in 14.8%, with one-third being due to overlying necrotic core producing attenuation of the echo signal. In the remaining two-thirds, only microcalcium deposits were evident on histology, which were not detected by IVUS.18

Using OCT to assess calcium offers a number of other advantages, including the ability to measure calcium thickness, arc and volume.26 Conversely, ultrasound waves are absorbed by calcium, resulting in an acoustic shadow limiting the ability to discern the depth and volume of calcium by IVUS.72 Broadly speaking, three morphological subtypes of calcium can be identified on intracoronary imaging: eccentric (calcium arc ≤180°), concentric (calcium arc >180°) and nodular (calcified protrusion into the vessel lumen) (Figure 3). Some studies have considered eccentric calcium as an arc of <270°.

Figure 3: Calcium Morphology by Both Optical Coherence Tomography and Intravascular Ultrasound.

A: OCT image demonstrating concentric calcium with an angle of 360° and thickness of 0.9 mm; B: OCT image demonstrating eccentric calcium with an angle of 110° and thickness of 0.8 mm; C: OCT image demonstrating a calcified nodule with a thick fibrous cap protruding into the lumen (white arrow); D: IVUS image demonstrating concentric calcium with an angle of 260°; E: IVUS image demonstrating eccentric calcium with an angle of 120°. Note the posterior acoustic shadowing (yellow asterisk) suggesting that the calcification is approximately 1 mm deep; F: IVUS image showing a protruding calcified nodule (yellow arrow). IVUS = intravascular ultrasound; OCT = optical coherence tomography.

Assimilation of other data gleaned from intracoronary imaging pullbacks can help to predict the risk of stent under-expansion if calcium is not appropriately modified. Fujino et al. derived a calcium-scoring system based on OCT known as the ‘rule of 5s.’ This scoring system attributed two points for a calcium arc >180° on OCT cross-section, one point for a calcium depth of >0.5 mm and one point for calcium length of >5 mm. A score of 4 was associated with stent expansion <80%.73 Similarly, Zhang et al. devised and validated an IVUS scoring system for predicting stent underexpansion.74 One point is attributed to each of the findings of calcium arc >270° for a length of >5 mm, 360° calcium within the lesion, presence of a calcified nodule within the lesion and vessel size <3.5 mm. A score of 2 or more demonstrates a risk of stent under-expansion (<70% expansion), which can be ameliorated using atherectomy techniques.74 The main utility of these scoring systems is to highlight the need for plaque modification prior to stenting. A target expansion of >80% is recommended in line with current consensus guidelines, although it is acknowledged that, despite the use of calcium modification techniques, this threshold is often difficult to reach in practice.71 Figure 4 demonstrates the minimum stent expansion achieved in contemporary studies using calcium modification techniques and intracoronary imaging.75–81 Minimum stent area (MSA) targets have been defined by previous studies demonstrating reduced TLF if reached.71 A MSA target by IVUS of 5.5 mm2 and 4.5 mm² by OCT are recommended by current guidelines. Target MSA is significantly larger for left main stem bifurcation stenting. An IVUS subgroup analysis of the NOBEL study demonstrated 0% TLR at 5 years in those with left main stem MSA ≥13.4 mm².82

Figure 4: Stent Expansion in Contemporary Studies of Calcium Modification and Intracoronary Imaging.

Note that many studies do not achieve the >80% expansion recommended in current consensus documents. *The study by Ali et al. examined stent expansion in different calcium arcs. Mean stent expansion at final MSA was 82.8% in calcium arc <180°, 78.4% in calcium arc 181–270°, 78.3% in calcium arc 271–359°and 79.1% in calcium arc 360°. IVL = intravascular lithotripsy; MB = modified balloon; MSA = minimum stent area; NCB = non-compliant balloon; OA = orbital atherectomy; RA = rotational atherectomy.

The non-compliant nature of calcified plaques means that modification and fracture of calcium sheets are required to allow stent expansion and achieve the target MSA.15 Many calcium modification techniques and devices are available, with varying mechanisms of action; however, head-to-head comparisons between these techniques are limited.83 Furthermore, the periprocedural complication profiles between devices used to modify calcium also vary. Small, single-arm observational studies and small randomised trials do exist and may provide some guidance for device selection for calcium modification. A number of calcium modification algorithms also exist, but none have been validated in a prospective manner; however, they act as a useful aide memoir for the cath lab. In the following section, we summarise the published data on the currently available calcium modification tools, along with presenting a pragmatic calcium modification algorithm (Figure 5). Further review articles in this series provide an in-depth analysis of each tool, together with their associated clinical data.84–87

Figure 5: Calcium Modification Algorithm Guided by Intracoronary Imaging.

CAD = coronary artery disease; IVL = intravascular lithotripsy; IVUS = intravascular ultrasound; LASER = light amplification by stimulated emission of radiation; MB = modified balloon; NCB = non-compliant balloon; OA = orbital atherectomy; OCT = optical coherence tomography; RA = rotational atherectomy.

Tools for Calcium Modification

Calcified lesions are often subdivided into those that are crossable with a device or uncrossable. Rotational atherectomy and excimer laser coronary atherectomy remain the predominant treatments in severely calcified uncrossable lesions.83 Crossable lesions may be treated using a variety of different tools. Balloon-based therapies are often initially attempted to determine the resistance ofthe plaque, and some studies exist comparing devices within this category. The recently published COPS study randomised 87 patients to either non-compliant balloon (NCB) or cutting balloon (CB) therapy.88 Larger final MSA by IVUS at the site of calcium was noted in those treated with a CB strategy (8.1 ± 2 mm² versus 7.3 ± 2.1 mm² for CB versus NCB balloon, respectively; p=0.035).88 Comparison of CBs and scoring balloons in a small retrospective study by Matsukawa et al. demonstrated greater acute gain, final MSA and stent symmetry with CB.89

Atherectomy techniques such as rotational atherectomy (RA) and orbital atherectomy (OA) are considered to be advanced calcium modification tools, which can be used as an upfront or bailout strategy if a lesion is not adequately dilated using a balloon-based therapy. RA compared with conventional therapy in the ROTAXUS trial resulted in initial greater procedural success and luminal gain but resulted in greater late luminal loss at 9-month angiographic follow-up.90 Furthermore, no differences in clinical outcomes at 2 years were seen between RA and conventional therapy.90,91

A more contemporary study of RA, the PREPARE-CALC study, compared RA with modified balloons (cutting and scoring), finding greater procedural success in the RA arm, predominantly driven by excess need for crossover to RA in the modified balloon arm (16% cross-over from modified balloon to RA).92 No differences in late lumen loss at 9 months were seen between groups.92 The efficacy of OA in treating calcified lesions has been demonstrated in a number of single-arm studies.93–95 More recently, the DIRO study compared RA and OA for calcified lesions (>180° calcium by OCT or moderate-to-severe by angiography).80 Both percentage luminal area increase (72.2% versus 39.2%; p<0.01) and stent expansion (99.5% versus 90.6%; p=0.02) were greater with RA compared with OA.80 Intravascular lithotripsy (IVL) is a novel calcium modification tool using acoustic waves to cause calcium fracture. Its safety and efficacy have been demonstrated in the DISRUPT CAD series of studies.96

Recently, our research group demonstrated the effectiveness of IVL in both eccentric (calcium arc ≤180°) and concentric (calcium arc >180°) calcification with no difference in MSA (5.9 mm² ± 2.2 mm² versus 6.25 mm² ± 2.4 mm²; p=0.483), minimum stent expansion (80.9% ± 16.7% versus 78.2 ± 19.8%), or stent expansion at the maximum calcium site (100.6% ± 24.2% versus 95.8 ± 27.3%) (p>0.05 for all comparisons of concentric versus eccentric, respectively) as assessed using OCT.21 This was despite greater fracture seen in concentric lesions.21 Additionally, while fractures within calcified nodules were rare, mean stent expansion at the nodule site was ~100%.21 A small, randomised study of 70 patients treated with either RA or IVL demonstrated no significant difference in final MSA between the two therapies, although MSA was numerically larger in the RA group.79 IVL, therefore, is becoming a useful tool in the management of all types of calcium morphologies.

Combination therapies are often required, and several studies have examined this. The PREPARE-CALC-COMBO prospective study examined the use of rotational atherectomy followed by cutting balloon and compared this to a historical cohort of RA or cutting balloon alone, finding greater luminal gain and MSA with the combination strategy.77 However, the randomised prospective ROTA-CUT trial found no differences in MSA between a combined RA-cutting balloon strategy compared with RA-non-compliant balloons.78 IVL has been used in a number of combinations; however, published data are limited and predominantly confined to case reports or small series. Sardella et al. reported a multicentre, prospective study of RA followed by IVL as either a planned or bailout strategy in 160 patients with low complication rates and freedom from in-hospital MACE of 98.7%.97 Combination therapies may be employed as an up-front or bailout strategy and we recommend interval intracoronary imaging after use of each calcium modification tool to assess for fracture and residual un-modified calcium that may affect stent expansion before employing our algorithmic stepwise approach (Figure 5). Finally, medical optimisation of co-morbidities contributing to the progression of coronary calcification is essential to maximise the longevity of the PCI result and overall patient outcome.

Conclusion

Calcified CAD is common and presents a number of difficulties when performing PCI. The presence of an atherosclerotic plaque is necessary for calcium deposition, which begins as microcalcifications and progresses to calcified sheets and nodular calcification. Patients with calcified CAD undergoing revascularisation experience poorer outcomes regardless of the mode of revascularisation (surgery or PCI). When performing PCI, stent under-expansion because of un-modified calcium can result in stent failure at both short- and long-term follow-up. Many calcium modification techniques exist, and calcium modification algorithms can assist in selecting devices for calcified lesions. Intracoronary imaging is essential when treating calcified CAD and can assist in device selection and optimisation of PCI results. Medical optimisation of contributing factors to coronary calcification is essential to optimise outcomes.