Access to side branches with a sharply angulated origin: usefulness of a specific wire for chronic occlusions

INTRODUCTION

Ischemic heart disease is the leading cause of disability and mortality worldwide and is commonly characterized by the presence of obstructive coronary artery disease (CAD) (defined as any coronary artery stenosis ≥ 50% in diameter).1 However, up to 60% to 70% of patients with angina and/or documented myocardial ischemia do not have angiographic evidence of CAD.2 This condition is defined as angina with no obstructed coronary arteries (ANOCA) or ischemia with no obstructed coronary arteries (INOCA) when associated with evidence of myocardial ischemia.3 Of note, despite the absence of CAD, these patients are at an increased risk of future cardiovascular events such as acute coronary syndrome, heart failure hospitalization, stroke, and repeat cardiovascular procedures compared with healthy individuals.4,5 Therefore, appropriate management in terms of diagnosis and treatment is of the utmost importance to improve patients’ prognosis and outcomes.6 The Coronary Microvascular Angina (CorMicA) trial demonstrated that a strategy of adjunctive invasive testing for disorders of coronary function together with stratified medical therapy can improve outcomes (ie, reduction in angina severity and enhanced quality of life).7,8 However, there are still concerns about the implementation in real-world practice of a systematic diagnostic and therapeutic approach in INOCA and ANOCA patients, potentially impacting outcomes and quality of life.

We report our single-center experience of the implementation in clinical practice of a specific diagnostic and therapeutic protocol (no obstructed coronary arteries [NOCA] program) in INOCA and ANOCA patients.

METHODS

Eligibility criteria for the NOCA program

All consecutive patients diagnosed either at our hospital or at our referral centers with angina or ischemia with nonobstructive CAD on coronary angiography were referred to a specific outpatient clinic (the NOCA clinic at Hospital Clínic, Barcelona, Spain) for a screening visit. Nonobstructive CAD was defined as angiographic evidence of normal coronary arteries or diffuse atherosclerosis with stenosis < 50% and/or fractional flow reserve (FFR) > 0.80 if there was stenosis between 50% and 70%. During the screening visit, a team of expert cardiologists confirmed patients’ eligibility for the NOCA program based on the following criteria: a) diagnosis of ANOCA, defined as stable, chronic typical angina symptoms (eg, chest pain precipitated by physical exertion or emotional stress and relieved by rest or nitroglycerine); b) diagnosis of INOCA, defined as the demonstration of myocardial ischemia identified by a noninvasive test with pharmacologic or exercise stress tests such as cardiac single photon emission computed tomography, cardiac magnetic resonance, stress electrocardiography, or echocardiography.3 The exclusion criteria were: a) atypical angina symptoms, and b) clearly identifiable noncoronary causes of chest pain (figure 1). The study protocol adhered to the Declaration of Helsinki and the study was approved by our institutional review committee. All patients provided written informed consent to be included in this program and study. The clinical ethics committee gave their approval for a retrospective analysis of the collected data.

Figure 1. Central illustration. Flowchart for the inclusion of patients in the NOCA program. Cath lab, catheterization laboratory; INOCA: ischemia with no obstructed coronary arteries; NOCA, no obstructed coronary arteries; NOCAD, nonobstructive coronary artery disease.

 

NOCA program: diagnostic approach

After patient inclusion in the NOCA program, specialized counseling was provided by expert cardiologists and nurses. All patients were thoroughly informed about their disease, the importance of reaching a specific diagnosis, and the importance implementing tailored therapy. During the counseling sessions, the predicted benefits and low associated risks of an invasive procedure to specifically study coronary microcirculation and vasospasm were explained in detail. All patients provided written informed consent to undergo coronary angiography and intracoronary provocation testing with acetylcholine (ACh).

Subsequently, all patients underwent a scheduled coronary angiogram with a comprehensive diagnostic work-up consisting of the following: a) coronary function testing to assess coronary flow reserve (CFR) and the index of microvascular resistance (IMR); b) intracoronary ACh provocation testing to assess the presence of coronary vasomotion disorders (eg, epicardial or microvascular spasm).

Coronary function testing was performed using a pressure-temperature sensor guidewire (PressureWire X Guidewire and Coroventis CoroFlow Cardiovascular System, Abbott Vascular, United States) placed in the left anterior descending artery (LAD) as the prespecified target vessel, reflecting its subtended myocardial mass and coronary dominance. Steady-state hyperemia was induced using intravenous adenosine (140 µg/kg/min). If there was severe tortuosity of the LAD or evidence of myocardial ischemia in a region other than the territory of the LAD, the wire was placed in the right coronary artery or the left circumflex, as per the operator’s decision. CFR was calculated using thermodilution, defined as resting mean transit time divided by hyperemic mean transit time (abnormal CFR was defined as ≤ 2.5). IMR was calculated as the product of distal coronary pressure at maximal hyperemia multiplied by the hyperemic mean transit time (normal value < 25).6,9

Intracoronary ACh provocation testing was performed with a standardized protocol involving serial ACh infusions for 20 seconds at increasing concentrations (2-20-100 µg in the left coronary artery with an interval of 2-3 minutes between each injection) with concomitant assessment of the patient’s symptoms, electrocardiogram documentation, and angiographic scans. Patients taking vasoactive drugs (eg, calcium channel blockers and nitrates) underwent a wash-out period of at least 48 hours before the provocative test.10,11,12 Epicardial coronary spasm was defined as the reproduction of chest pain and ischemic electrocardiogram changes in association with a reduction in coronary diameter ≥ 90% from baseline in any epicardial coronary artery segment.13 Microvascular spasm was diagnosed when typical ischemic ST-segment changes (deviation ≥ 1 mm) and angina developed in the absence of epicardial coronary constriction (< 90% diameter reduction).14

Subsequently, patients were stratified into 4 endotypes: a) microvascular angina (MVA) (evidence of coronary microvascular dysfunction [CMD] defined as any abnormal CFR [< 2.5], IMR [≥ 25], or microvascular spasm); b) vasospastic angina (VSA) (CFR ≥ 2.5, IMR < 25 and epicardial spasm); c) both MVA and VSA (evidence of CMD and epicardial spasm); and d) noncoronary chest pain (CFR ≥ 2.5 and IMR < 25, with neither microvascular nor epicardial spasm).6

Any complications occurring during the invasive diagnostic work-up were documented, including bradyarrhythmias, atrial fibrillation, ventricular tachycardia or fibrillation, coronary perforations, death from any cause, and any other complications.

NOCA program: pharmacological and psychological therapeutic approach

Once the endotype was identified, medical treatment for each patient was optimized accordingly (table 1). In patients with MVA, treatment with beta-blockers and calcium channel blockers (CCBs) was started or up-titrated. Ranolazine was added if angina symptoms were not fully controlled by beta-blockers and CCBs. In patients with VSA, treatment with nondihydropyridine CCBs and long-acting nitrates was started or up-titrated. In patients with both MVA and VSA, treatment with nondihydropyridine CCBs or beta-blockers was started or up-titrated. In patients with noncoronary chest pain, vasoactive drugs were discontinued unless clinically indicated for other reasons. Additionally, treatment with angiotensin-converting enzyme inhibitors/angiotensin receptor blockers and statins was started or up-titrated in all patients. If a patient showed intolerance or had contraindications to a specific medication (eg, asthma for beta-blockers, perimalleolar edema for CCBs, severe bradycardia for both beta-blockers and CCBs), the treatment was tailored and modified accordingly.

Because stress is an important trigger factor for angina symptoms, all patients were also referred to a team of expert psychologists for psychological support.15

NOCA program: clinical outcome and quality of life evaluation

All patients were followed up at 1, 3, 6, and 12-months for treatment titration and assessment of clinical outcomes. At the time of coronary angiography (ie, baseline) and at the 3-month follow-up, all patients were administered the Seattle Angina Questionnaire (SAQ) and quality of life questionnaire (EuroQol-5D [EQ-5D]). The SAQ is a validated 19-item self-administered questionnaire that measures 5 dimensions of CAD: physical limitation, angina stability, angina frequency, treatment satisfaction, and disease perception.16 The EQ-5D is a standardized, nondisease-specific questionnaire used to describe and evaluate patients’ health status and was intended to complement other quality-of life measures.17 Figure 2 provides a visual representation of all the steps involved for patients included in the NOCA program.

Figure 2. Visual representation of the NOCA program. EQ-5D, EuroQol-5D; NOCA, no obstructed coronary arteries; SAQ, Seattle Angina Questionnaire. * See text for more details.

 

Statistical analysis

Data distribution was assessed according to the Kolgormonov-Smirnov test. Continuous variables were compared using the unpaired Student t-test or the Mann–Whitney U test, as appropriate. The data are expressed as mean ± standard deviation (SD) or as median and interquartile range [IQR]. Categorical data are expressed as numbers and percentages and were evaluated using the chi-square test or Fisher exact test, as appropriate. A 2-sided P value < .05 was considered significant. All analyses were performed using SPSS version 21 (SPSS, United States).

RESULTS

Baseline characteristics of the study population

From January 2021 to December 2021, a total of 77 patients were screened at the NOCA clinic for inclusion in the NOCA program. Following the screening visit, 23 (29.9%) patients were excluded from the NOCA program: 12 due to atypical angina symptoms and 11 due to a clearly identifiable noncoronary cause. Consequently, 54 patients were included in the NOCA program (mean age 64.4 ± 9.4 years, 39 [63.9%] women). A total of 29 (53.7%) patients had INOCA and 25 (46.3%) had ANOCA. All clinical and angiographic characteristics of the study population are shown in table 1.

 

Table 1. Medical therapy according to the specific endotype of ANOCA/INOCA

Pathogenic mechanism of MINOCA	Therapeutic implications
MVA	Beta-blockers (Nebivolol 2.5–10 mg daily)
	CCBs (amlodipine 10 mg daily, or verapamil 240 mg daily, or diltiazem 90 mg twice daily)
	Ranolazine (375-750 mg twice daily)
VSA	Nondihydropyridine CCBs (verapamil 240 mg, or ciltiazem 90 mg twice daily)
	Long-acting nitrates (isosorbide mononitrate 30 mg)
MVA and VSA	CCBs (verapamil or diltiazem) or beta-blockers
Noncoronary chest pain	Beta-blockers or dihydropyridine CCBs if clinically indicated (eg, hypertension)
	ACEi or ARB if clinically indicated
	Statins if clinically indicated
ACEi, angiotensin-converting enzyme inhibitors; ANOCA, angina with no obstructed coronary arteries; ARB, angiotensin receptor blockers; CCBs, calcium channel blockers; INOCA, ischemia with no obstructed coronary arteries; MINOCA, myocardial infarction with non-obstructive coronary artery disease; MVA, microvascular angina; VSA, vasospastic angina.

 

NOCA program: diagnosis of the specific endotype and complications

The results of the invasive functional assessment are presented in table 2. The mean IMR and CFR values were 21.2 ± 10.6 and 2.3 ± 1.4, respectively. MVA was diagnosed in 19 (35.2%) patients, VSA in 12 (22.2%), and both MVA and VSA in 18 (33.3%). Finally, 5 (9.3%) patients were diagnosed with noncoronary chest pain.

 

Table 2. Clinical and angiographic characteristics of patients included in the NOCA program

Characteristics	Study population (n = 54)
Clinical characteristics
Age	64.4 ± 9.4
Female sex	39 (72.2)
Clinical presentation
ANOCA	25 (46.3)
INOCA	29 (53.7)
Diabetes mellitus	12 (22.2)
Hypertension	35 (64.8)
Dyslipidaemia	28 (51.9)
Former smokers	3 (5.7)
Current smoker	14 (25.9)
Familiar history of CV disease	5 (9.3)
Previous CV history
Prior MI	7 (13.0)
Prior PCI	8 (14.8)
Prior CABG	0 (0.0)
COPD	1 (1.9)
CKD (eGFR < 60 mL/min/m2)	4 (7.4)
Depression	15 (27.8)
Anxiety	19 (35.2)
Invasive functional evaluation
Vessel explored
LDA	48 (88.9)
LCx	3 (5.6)
RCA	3 (5.6)
IMR	21.2 ± 10.6
Increased IMR (≥ 25)	18 (33.3)
CFR	2.3 ± 1.4
Reduced CFR (< 2.5)	33 (61.1)
Increased IMR (≥ 25) and reduced CFR (< 2.5)	13 (24.1)
Diagnosis (endotype)
MVA	19 (35.2)
VSA	12 (22.2)
MVA and VSA	18 (33.3)
Noncoronary chest pain	5 (9.3)
ANOCA, angina with no obstructed coronary arteries; CABG, coronary artery bypass graft surgery; CFR, coronary flow reserve; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CV, cardiovascular; eGFR, estimated glomerular filtration rate; IMR, index of microcirculatory resistance; INOCA, ischemia with no obstructed coronary arteries; MI, myocardial infarction; MVA, microvascular angina; LAD, left anterior descending; LCx, left circumflex; PCI, percutaneous coronary intervention; RCA, right coronary artery; VSA, vasospastic angina. Values are expressed as No. (%), mean ± standard deviation or median [interquartile range].

 

Among INOCA patients, MVA was diagnosed in 11 (37.9%) patients, VSA in 7 (24.1%), both MVA and VSA in 8 (27.6%), and noncoronary chest pain in 3 (10.3%). Among ANOCA patients, MVA was diagnosed in 8 (32.0%) patients, VSA in 5 (20.0%), both MVA and VSA in 10 (40.0%), and noncoronary chest pain in 2 (8.0%). There were no statistically significant differences in the prevalence of any endotype between INOCA and ANOCA patients (all P > .05, figure 3).

Figure 3. Prevalence of the different endotypes among INOCA and ANOCA patients. ANOCA, angina with no obstructed coronary arteries; INOCA, ischemia with no obstructed coronary arteries; MVA, microvascular angina; NCCP, noncoronary chest pain; VSA, vasospastic angina.

 

Complications occurred in 3 (5.5%) patients during intracoronary ACh provocation testing: 2 (3.7%) patients had transient bradyarrhythmias and 1 (1.8%) patient had paroxysmal atrial fibrillation that spontaneously reverted to sinus rhythm.

NOCA program: treatment optimization according to the specific endotype

Inclusion in the NOCA program led to statistically significant changes in medications after diagnosis of the specific endotype. There was a significant increase in the use of beta-blockers (33.3% before vs 57.4% after, P = .008), nondihydropyridine CCBs (9.3% before vs 37.0% after, P < .001), and long-acting nitrates (46.3% before vs 63.0% after, P = .012). There were no statistically significant differences in any other medications before and after the invasive assessment (all P > .05, figure 4). All changes in medications according to the specific endotype of ANOCA/INOCA are shown in figure 5.

Figure 4. Differences in medical treatment before and after patient inclusion in the NOCA program. ACEi, angiotensin-converting enzyme inhibitors; ARB, angiotensin receptor blockers; DP-CCBs, dihydropyridine calcium channel blockers; LA, long-acting; ND-CCBs, nondihydropyridine calcium channel blockers.

Figure 5. Changes in medications according to the specific endotype of ANOCA/INOCA. A: patients with MVA. B: patients with VSA. C: Patients with MVA and VSA. D: patients with noncoronary chest pain. DP-CCBs, dihydropyridine calcium channel blockers; LA, long-acting; MVA, microvascular angina; ND-CCBs, nondihydropyridine calcium channel blockers; VSA, vasospastic angina.

 

NOCA program: clinical outcome evaluation

At 3 months of follow-up, there was no statistically significant difference in the mean EQ-5D score compared with baseline (64.8 ± 18.1 at baseline vs 66.1 ± 17.1 at 3 months of follow-up, P = .302) (figure 6). However, there was a statistically significant improvement in the SAQ score in terms of physical limitations (59.7 ± 19.3 at baseline vs 66.2 ± 16.9 at 3 months of follow-up, P = .037), angina stability (57.1 ± 28.1 at baseline vs 75.8 ± 22.3 at 3 months of follow-up, P = .010), and disease perception (42.5 ± 13.9 at baseline vs 50.8 ± 16.3 at 3 months follow-up, P = .015). No statistically significant difference was found in angina frequency (74.3 ± 20.4 at baseline vs 80.7 ± 19.8 at 3 months of follow-up, P = .193) or treatment satisfaction (68.1 ± 12.6 at baseline vs 70.5 ± 12.5 at 3 months of follow-up, P = .950) (figure 7). No events were recorded at the 1-year follow-up.

Figure 6. Differences in EuroQol-5D (EQ-5D) score at baseline and at 3 months of follow-up.

Figure 7. Differences in SAQ score at baseline and at 3 months of follow-up. AF, angina frequency; AS, angina stability; DP, disease perception; EQ-5D, EuroQol-5D; PL, physical limitations; SAQ, Seattle Angina Questionnaire; TS, treatment satisfaction.

 

DISCUSSION

The main results of our experience can be summarized as follows: a) the implementation of a specific diagnostic and therapeutic protocol (NOCA program) in patients with diagnosed with nonobstructive CAD is feasible and allowed a parsimonious use of medical resources; b) a comprehensive diagnostic work-up in INOCA and ANOCA patients is safe, with a low rate of mild and transient complications (5.5%); c) the inclusion of patients in the NOCA program led to significant changes in medications and a significant improvement in their angina symptoms at the 3-month follow-up with no adverse events at 1 year.

Although accumulating evidence has demonstrated that an approach consisting of a comprehensive diagnostic assessment and stratified medical therapy in INOCA and ANOCA is crucial to improve patients’ prognosis, such an approach is far from routinely implemented in clinical practice.7,8 There are still concerns mainly related to the cost-benefit ratio, the associated prolonged procedural time, increased costs, and the risk of possible associated complications. Furthermore, in the most recent European Society of Cardiology guidelines, invasive coronary function testing is assigned a class IIa (“should be considered”) recommendation, while ACh provocation testing is supported by a class IIb recommendation (“may be considered”) to assess microvascular spasm and class IIa in patients under consideration for VSA.3 As a result, the management of these patients is commonly left to physicians’ discretion or relies on the experience of each center. Consequently, diagnosis of a specific NOCA endotype is frequently missed, and medical therapy is not optimized. This, in turn, has a significant negative impact on patients’ quality of life and clinical outcomes, as well as on health care costs, due to the need for repeat hospitalization or invasive procedures.18

In reporting our experience, we demonstrate that a specific diagnostic and therapeutic protocol (ie, the NOCA program) in patients with a previous diagnosis of nonobstructive CAD can be easily implemented in clinical practice. A key innovation of our study, compared with prior publications, is the creation and implementation of a specific protocol for the INOCA/ANOCA population. Additionally, our approach involves a screening visit with assessment by a team of expert cardiologists for patients with a suspected diagnosis of INOCA/ANOCA. This approach improves identification of such patients, and, in our experience, led to the exclusion of almost one third of patients (29.9%) due to atypical angina symptoms or no clearly identifiable coronary causes of chest pain. This is another novelty of our study that could be extremely relevant in the management of these patients. Indeed, the selection of patients to be included in the program may allow clinical resources to be directed to patients who are most likely to benefit, while avoiding repeat invasive procedures and related risks in patients with unclear indications. Additionally, the specialized counseling provided by cardiologists and nurses during the screening visit, together with psychological support, are likely to be vital components of the management of INOCA/ANOCA patients. Indeed, recent studies have demonstrated how psychological factors, such as chronic stress, anxiety, depression, and social stressors are involved in the pathogenesis of MVA and VSA.19-23 Mental stress has been demonstrated to determine CMD mainly through endothelium-dependent mechanisms and endothelial dysfunction.24 Similarly, by activating brain areas involved in regulation of neuroendocrine and autonomic nervous systems, mental stress can lead to hyperreactivity of vascular smooth muscle cells, autonomic nervous system dysfunction, oxidative stress, vascular inflammation, and endothelial dysfunction, resulting in an increased propensity to coronary vasospasm.25-27

Furthermore, in line with previous studies,28-30 our experience demonstrates that performing a comprehensive invasive diagnostic assessment for the diagnosis of the specific endotype in INOCA and ANOCA patients is safe and is associated with a low rate of mild and transient complications. For all these reasons, patients and clinicians should be reassured about the lack of serious complications and cardiologists should be strongly encouraged to implement a specific diagnostic and therapeutic program in these patients. Indeed, the availability of such a program for INOCA and ANOCA patients may have significant clinical and therapeutic implications, as, in our experience, it resulted in substantial changes in medications and a marked improvement at the 3-month follow-up of the SAQ questionnaire regarding physical limitations, angina frequency, and disease perception. The lower and nonsignificant improvement in the other parameters (eg, angina frequency and treatment satisfaction) could be attributed to the already high baseline values (74.5 ± 19.9 and 69.6 ± 11.9, respectively). Similarly, the absence of a significant improvement in the EQ-5D questionnaire at 3 months might be due to the short follow-up period or the fact that it is a nondisease-specific questionnaire designed to describe and assess patients’ health status and is intended to complement other quality-of-life measures.31

Study limitations

Some limitations of this study should be acknowledged. First, this is a single-center study with a relatively small sample size and short follow-up. Second, we did not perform a cost-analysis and therefore we cannot speculate on the impact of the NOCA program on health care-related costs. Further studies in larger ANOCA and INOCA populations are warranted. Finally, the absence of a control group precluded a thorough assessment of the improvement in the quality of life among these patients.

CONCLUSIONS

Our experience demonstrates that a specific diagnostic and therapeutic protocol (NOCA program) can be easily and safely implemented in routine clinical practice. Such a protocol could ensure the best care for INOCA and ANOCA patients, as well as improve their quality of life and avoid inappropriate treatments and incomplete investigations. Future evidence from randomized clinical trials or recommendations from international clinical guidelines supporting the implementation of a specific protocol in these patients are strongly warranted.

FUNDING

This study received no funding.

ETHICAL CONSIDERATIONS

The study protocol complied with the Declaration of Helsinki and the study was approved by our Institutional Review Committee. All patients gave written informed consent to be included in this program and study. The clinical ethics committee gave their approval for a retrospective analysis of the data collected. In this work, the possible variables of sex and gender have been taken into account.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

No artificial intelligence tools were used during the preparation of this work.

AUTHORS’ CONTRIBUTIONS

R. Rinaldi, F. Spione, F.M. Verardi: data extraction and analysis and manuscript drafting; R. Rinaldi, F. Spione, S. Brugaletta: design and manuscript revision; P. Vidal Calés, V. Arévalos, R. Gabani, D. Cánovas, M. Gutiérrez, M. Pardo, R. Domínguez, L. Pintor, X. Torres, X. Freixa, A. Regueiro, O. Abdul-Jawad Altisent, M. Sabaté: manuscript revision. All authors have read and agreed to the published version of the manuscript.

CONFLICTS OF INTEREST

The authors have nothing to disclose.

ACKNOWLEDGMENTS

F. Spione has been supported by a research grant provided by the Cardiopath PhD program.z

REFERENCES

INTRODUCTION

Moderate or severe coronary artery calcification is relatively common in patients undergoing percutaneous coronary interventions (PCI).1 This is closely related to advancing age and the high prevalence of comorbidities such as diabetes or chronic kidney disease. Coronary artery calcification is associated with a lower rate of successful PCI and complete revascularization, increased procedural-related complications, and a higher rate of major adverse cardiovascular events (MACE).2

Despite the availability of several plaque modification techniques, severely calcified lesions continue to pose a challenge to the successful performance of PCI.

Excimer laser coronary atherectomy (ELCA) is a plaque modification technique that has proved to be useful in various scenarios such as balloon failure (uncrossable or undilatable lesions), chronic total occlusions (CTO), stent underexpansion, in-stent restenosis (ISR), and thrombotic lesions. In recent years, incremental operator experience along with the standardization of the laser technique, has expanded its indications and decreased its complication rates.3,4

However, its effectiveness in calcified lesions is controversial. On one hand, some ELCA series have described a relationship between severe calcification and laser failure.5-8 On the other hand, moderate-to-severe calcification is found in more than 60% of cases in some ELCA series with a high success rate,9 suggesting that it could be useful in this setting.10

Due to the lack of evidence in this specific scenario, the aim of our study was to assess the safety and efficacy of ELCA in severely calcified coronary lesions, as well as the mid-term follow-up outcomes in a single center registry.

METHODS

Patient population

This single center retrospective observational study included all consecutive patients undergoing ELCA-assisted PCI for severely calcified lesions. From March 2016 to August 2022.

We excluded procedures using ELCA for any indication other than severe calcification. In all patients, PCI was indicated based on the presence of symptoms consistent with angina, demonstrated ischemia, or both. The study followed the international recommendations of clinical investigation (Declaration of Helsinki). All participants gave written informed consent and approval was obtained from the ethics committee of the center. The study took into consideration sex and gender variables according to SAGER guidelines. Patients were followed up in cardiology clinics at their referral center 3 to 6 months after the procedure, and thereafter at time intervals established at the discretion of their treating physician.

We analyzed data on clinical and angiographic characteristics, technical aspects of the procedure, and cardiovascular events during hospitalization and after discharge.

Procedure

All procedures were carried out by 5 different operators experienced in the use of ELCA. The decision to use ELCA was based on the presence of angiographically severe calcification.

Radial access was use by default. All cases were performed with the CVX-300 Excimer Laser System (Philips, Netherlands) using the 0.9 mm or 1.4 mm catheters. Saline infusion technique was used by default from the beginning, with fluence (mJ/mm²), frequency or repetition rate (Hertz), and the possibility to use ELCA without saline infusion or even with contrast left to the operator’s discretion. Additional dilatation with noncompliant balloons was performed in all procedures. Patients in which another plaque modification technique was used in combination with ELCA were included. All PCIs were performed following current recommendations.11

Definitions

Severely calcified lesions were angiographically defined as radiopacities observed on fluoroscopy without cardiac motion before contrast injection compromising 1 or both sides of the lumen.12 Balloon-uncrossable lesions were defined as lesions that could not be crossed with the lowest-profile balloon available or a microcatheter despite successful advancement of the guidewire distal to the lesion, having good guide catheter support with a guide extension catheter when required. Balloon-undilatable lesions were defined as those lesions in which a noncompliant balloon (diameter 1:1 according to the vessel diameter) failed to achieve adequate expansion. Anterograde flow was assessed by the Thrombolysis In Myocardial Infarction (TIMI) scale.

ELCA success was defined as the ability to cross the lesion completely with the laser catheter or, if crossing was not feasible, to allow the subsequent passage and expansion of a balloon sized 1:1 with the vessel diameter, after laser application. Technical success was defined as residual stenosis < 30% and anterograde TIMI 3 flow in the target vessel. Clinical success was defined as technical success and the absence of MACE during the current hospitalization (target lesion revascularization, procedure-related myocardial infarction [MI], or cardiovascular death). Procedural-related complications included coronary artery perforation leading to cardiac tamponade and requiring pericardial drainage, flow-limi­ting dissection, no-reflow, hemodynamic instability, MI type 4a according to the fourth universal definition of MI,13 ventricular arrhythmias, and major bleeding (bleeding requiring transfusion and/or surgical or percutaneous intervention). MACE occurring during follow-up were defined as a composite of target lesion revascularization, MI, or cardiac death.

Statistical analysis and data collection

All data were collected through the patients’ electronic medical records and were introduced in a local database. Angiograms were evaluated using local quantitative coronary analysis software and visual operators’ assessment. Categorical variables are reported as absolute values and percentages. Continuous variables are presented as the mean ± standard deviation (SD) or median (interquartile range [IQR] 25-75), depending on their normal or nonnormal distribution. All analyses were performed with StatIC 16.1 statistical software package.

RESULTS

Clinical characteristics

During the study period, a total of 78 patients with severely calcified coronary lesions underwent 80 ELCA-assisted PCIs and were included in the analysis. Patients undergoing ELCA for an indication other than severe calcification were excluded from the analysis. The distribution of the number of procedures per year, between March 2016 and May 2022, is shown in figure 1. A flowchart of patients in the present study is summarized in figure 2. Mean age was 71.2 ± 8.6 years, 62 (80.5%) were men, and there was a high prevalence of cardiovascular risk factors. Mean left ventricle ejection fraction was 52.9% ± 12.5%. Thirty-nine patients (50%) had a previous PCI. Clinical presentation was stable coronary artery disease in 45 procedures (56.2%), non–ST-segment elevation MI (NSTEMI) in 28 (35%), and ST-segment elevation MI (STEMI) in 7 (8.8%). Baseline clinical characteristics are summarized in table 1.

Figure 1. Distribution of the number of procedures per year (March 2016-May 2022).

Figure 2. Flowchart of patients in the present study. ELCA, excimer laser coronary atherectomy; PCI, percutaneous coronary intervention; RA, rotational atherectomy.

Table 1. Baseline clinical characteristics

Age	71.2 ± 8.6
Male sex	62 (80.5%)
Body mass index (kg/m2)	28.7 ± 4.2
Hypertension	70 (89.7%)
Dyslipidemia	61 (78.2%)
Diabetes mellitus	46 (59.0%)
Current smoker	19 (24.4%)
Prior PCI	39 (50.0%)
Prior CABG	8 (10.3%)
Hb (g/dL)	13.5 ± 5.3
Serum creatinine (mg/dL)	1.42 ± 1.8
Ejection fraction (%)	52.9% ± 12.5
Clinical presentation (n = 98)
Stable coronary artery disease	45 (56.2%)
NSTEMI	28 (35.0%)
STEMI	7 (8.8%)
CABG, coronary artery bypass graft surgery; NSTEMI, non–ST-segment elevation myocardial infarction; PCI, percutaneous coronary intervention; STEMI, ST-segment elevation myocardial infarction. Data are expressed as no. (%) or mean ± standard deviation.

 

Angiographic characteristics

Severe multivessel disease was present in 56 patients (71.8%). The most common target vessel was the left anterior descending artery (38.75%). In 7 procedures (8.75%), more than 1 target vessel were identified. The anatomical settings in the target vessel included uncrossable lesions in 32 (40%), undilatable lesions in 23 (28.75%), ISR in 2 (2.5%), and stent underexpansion related to calcified plaque in 5 (6.25%). In 6 (7.5%) procedures, the main indication for ELCA was CTO combined with any of the previous settings. In 10 procedures (12.5%), the indication for ELCA resulted from the combination of 2 or more of the above. ELCA was used with the sole indication of severely calcified lesion, not included in any of the previous anatomical settings, in 2 procedures (2.5%).

Procedural characteristics

The radial approach was performed in 44 (55%) cases. There was no need for access conversion when the radial approach was attempted.

Dual antiplatelet treatment consisted of pretreatment with aspirin and oral P2Y₁₂ receptor blockers in 58 patients (72.5%). Selection of P2Y₁₂ inhibitor was left to the physician’s discretion. Cangrelor was used in the patients without prior dual antiplatelet treatment. After the procedure and during follow-up, dual antiplatelet treatment was prescribed as follows: in stable coronary artery disease (n = 45) clopidogrel was used in 21 patients, ticagrelor in 10 and prasugrel in 3 patients. In acute settings (n = 35), ticagrelor was administered in 16 patients, prasugrel in 10, and clopidogrel in 7. GPIIB/IIIA inhibitors were used in 6 procedures (7.5%) (tirofiban in all cases).

Intracoronary imaging was used in 58 procedures (72.5%). Optical coherence tomography (OCT) was used in 48 procedures (60%) and intravascular ultrasound in 10 (12.5%).

Circulatory support with intra-aortic balloon counterpulsation was required in only 1 patient in the context of left-main revascularization.

Regarding the ELCA technique, most lesions were treated with 0.9 mm laser catheters (97.5%). In 2 patients, larger catheters (1.4 mm) were used (1 case of ISR in the left anterior descending artery and 1 calcified lesion in a saphenous vein graft). Flushing saline was used in all the procedures, and contrast was required in 2 procedures (figure 3). Maximum fluence was 73 ± 9.6 mJ/mm² and the maximum frequency 72.7 ± 10.4 Hz. The highest fluence of 80 mJ/mm² was required in 48 (60%) procedures and the highest frequency of 80 Hz in 48 (60%). A mean of 5103 ± 3120 pulses was delivered, and the median lasing time was 62 seconds [IQR 40-91].

Figure 3. In-stent restenosis and stent underexpansion treated by excimer laser coronary atherectomy (ELCA). Severe in-stent restenosis (ISR) (A1) (arrow) of drug-eluting stent previously implanted in the right coronary artery. Optical coherence tomography (OCT) revealed calcified neoatherosclerosis with a minimum luminal area (MLA) of 1.25 mm² (A2). An everolimus-eluting stent (2.75 × 20 mm) was implanted, and despite postdilatation with a 3-mm noncompliant (NC) balloon (A3), subsequent OCT confirmed stent underexpansion (MLA: 2.1 mm²) (A4). Sixteen months later, critical ISR of the previous stent (B1) (arrow) was noted with heterogeneous neointimal proliferation (B2). Laser atherectomy was performed, followed by dilation with 3- and 3.5-mm NC balloons up to 24 atm, and a 3-mm sirolimus-eluting stent was implanted with acceptable angiographic expansion (B3) but underexpansion on OCT (MLA: 1.5 mm²) (B4) (arrow). Laser application with contrast injection was repeated and was dilated with a 4 mm NC balloon, achieving adequate stent expansion (MLA: 4.5 mm²) (B5, B6) (arrow).

 

At least 1 new-generation drug-eluting stent was implanted in 70 procedures (87.5%). In the remaining procedures, stents were not delivered because of the presence of previous stents (6 ISR and 2 cases of stent underexpansion), which were treated with noncompliant and/or drug-eluting balloons, or due to ELCA failure (2 cases).

Angiographic and procedural characteristics and procedural strategy data are summarized in table 2.

 

Table 2. Angiographic and procedural characteristics

Angiographic characteristics
Target vessel
Left anterior descending coronary artery	31 (38.75%)
Right coronary artery	28 (35.0%)
Circumflex artery	10 (12.5%)
Left main coronary artery	4 (5.0%)
Multivessel disease	56 (71.8%)
Indication for ELCA
Balloon-uncrossable lesion	32 (40%)
Balloon-undilatable lesion	23 (28.75%)
In-stent restenosis	2 (2.5%)
Stent Underexpansion	5 (6.25)
Chronic total occlusion	6 (7.75%)
Combination of > 2 of the above	10 (12.5%)
Severe calcification as sole indication	2 (2.5%)
Bifurcation	14 (17.7%)
Aorto-ostial	2 (2.5%)
Procedural characteristics
Access site
Radial	44 (55.0%)
Femoral	33 (41.2%)
Femoral-radial	3 (3.8%)
Guiding catheter French
6-Fr	40 (50.0%)
7-Fr	34 (42.5%)
Intracoronary imaging	58 (72.5%)
OCT	48 (60.0%)
IVUS	10 (12.5%)
Laser catheter
1.4 mm rapid-exchange catheter	2 (2.5%)
0.9 mm rapid-exchange catheter	78 (97.5%)
Maximum fluence (mJ/mm2)	72.97 ± 9.6
Maximum frequency (Hz)	72.7 ±10.4
Number of pulses	5 103 ± 3 120
Total lasing time (sec)	62 [40-91]
Contrast volume (mL)	211 ± 68.0
Fluoroscopy time (min)	30 [22-39]
Radiation dose (Gy/cm2)	103 [79-185]
Procedural time (min)	72 [55-100]
Stent implantation	70 [87.5%]
Stent diameter (mm	3.04 ± 0.50
Stents per procedure	1.8 ± 1.14
Total stent length (mm)	43.7 ± 25.7
Left ventricle assist device used	1 (1.25%)
Timing of PCI (n = 98)
Ad hoc	22 (27.5%)
Deferred	58 (72.5%)
ELCA, excimer laser coronary atherectomy; OCT, optical coherence tomography; IVUS, intravascular ultrasound; PCI, percutaneous coronary intervention. Data are expressed as no. (%), mean ± standard deviation or median [interquartile range].

 

Procedural outcomes

The ELCA success rate was 91.25%. The success rate was 78.1% in uncrossable lesions and 100% in the other anatomical settings (P < .001). The ELCA success rate in the different anatomical settings is shown in figure 4.

Figure 4. ELCA success rate in the different anatomical settings. ISR, in-stent restenosis, CTO, chronic total occlusion.

 

Among intracoronary imaging-guided procedures, the ELCA success rate was 98.3%, and dropped to 72.7% in non-coronary imaging-guided PCI (P < .001). Final stent expansion was analyzed with intracoronary imaging in 32 procedures. The median stent expansion was 80.3% [IQR, 68.2%-95.2%].

Despite ELCA success, adjuvant plaque modification therapies (other than noncompliant [NC] balloon inflation after ELCA) were used in 4 procedures, including rotational atherectomy (RA) in 2 procedures, lithotripsy in 1 procedure and scoring balloon in 1 procedure. The procedures in which ELCA allowed subsequent successful RA (RASER technique14) or successful lithotripsy (ELCA-tripsy technique15) were considered ELCA success.

In 7 procedures (8.75%), ELCA failed. In 2 of them, RA was successfully performed. In 1 procedure, intravascular lithotripsy was attempted, but failed. In 1 case, the procedure was prematurely interrupted at the request of the patient. In the remaining 2 patients, no bailout therapy was attempted, and they were managed conservatively. Cases in which ELCA did not facilitate the passage of RA or intravascular lithotripsy were not classified as RASER or ELCA-tripsy techniques. The overall technical success rate was 91.25%.

In-hospital and follow-up outcomes

ELCA-related complications occurred in 2 procedures (2.5%) due to coronary artery perforation after ELCA application, with immediate sealing after stent implantation (although pericardiocentesis was necessary in 2 of them). A third perforation was observed, not immediately after ELCA application, but after dilatation with NC balloons. In 2 of the perforations, the target lesion was a severely calcified and undilatable lesion located in the left anterior descending artery. The third perforation was observed in an uncrossable lesion at the right coronary artery. In all of them, the 0.9 mm catheter was used, and ELCA was applied with maximum fluency and repetition rate during saline infusion. Intracoronary imaging prior to ELCA application was not performed in any of these patients: the OCT catheter could not cross the lesion in 2 of them and crossing was not attempted in the third. After the application of coronary laser and stent implantation, OCT was performed in 2 of the procedures, which confirmed the good final result.

Other procedural complications not related to ELCA occurred in 4 patients. One patient developed a vascular access complication with retroperitoneal hemorrhage and severe bleeding requiring transfusion and transarterial embolization of a deep femoral artery branch, although his clinical course was favorable. One patient with severe aortic stenosis and impaired left ventricular function showed hemodynamic instability requiring support with inotropes and orotracheal intubation. In 1 patient, no-reflow phenomenon occurred after stent implantation but resolved after intracoronary adenosine infusion.

In the remaining patient, coronary dissection occurred during the guidewire advancement before ELCA application and was complicated with occlusive intracoronary hematoma, which resolved after emergent PCI with successful revascularization. No patient died during the procedure. Three patients died during admission despite successful revascularization due to cardiac causes not related to the procedure (mostly advanced heart failure) and 1 patient died from respiratory sepsis. There were no other in-hospital complications. Overall, the clinical success rate was 87.5%.

After a median follow-up of 15.5 months [IQR, 5.02-29.3], MACE occurred in 9 patients (11.25%). Target lesion revascularization occurred in 7 patients (8.9%), in all patients due to ISR. The median time to target lesion revascularization among patients with successful revascularization was 11.4 [IQR, 8.1-22.6] months. Cardiorespiratory arrest secondary to acute stent thrombosis occurred in 1 patient with successful revascularization, whose family reported poor antiplatelet therapy adherence. One patient died from advanced heart failure after 3 years of follow-up, despite successful revascularization. Three patients died from noncardiac causes.

The procedural outcomes, clinical outcomes, and major complications are summarized in table 3. No significant differences were observed in the results between male and female patients.

 

Table 3. Procedural and clinical outcomes

Procedural and clinical success	n (%)
ELCA success	73 (91.25%)
Balloon-uncrossable lesion	25 (78.13%)
Balloon-undilatable lesion	23 (100%)
In-stent restenosis	2 (100%)
Stent underexpansion	5 (100%)
Chronic total occlusion	6 (100%)
Combination of > 2 of the above	10 (100%)
Severe calcification as sole indication	2 (100%)
Technical success	73 (91.25%)
Clinical success	70 (87.5%)
Procedural complications
ELCA-related complications
Coronary artery perforation	2 (2.5%)
Complications not related to ELCA
Vascular access complication with major bleeding	1 (1.25%)
Coronary perforation	1 (1.25%)
Flow-limiting dissection	1 (1.25%)
Hemodynamic instability	1 (1.25%)
No-reflow	1 (1.25%)
Ventricular arrhythmia	0 (0%)
In-hospital MACE
Recurrent angina requiring TLR	0 (0%)
Procedure-related myocardial infarction	1 (1.25%)
New-onset heart failure	0 (0%)
Stroke	0 (0%)
Cardiovascular death	3 (3.75%)
All-cause death	4 (5.0%)
MACE after discharge
TLR	7 (8.75%)
MI due to stent thrombosis	1 (1.25%)
Death from cardiovascular causes	2 (2.5%)
Non-cardiovascular related death	3 (3.75%)
ELCA, excimer laser coronary atherectomy; MACE, major adverse cardiovascular events; MI, myocardial infarction; TLR, target lesion revascularization.

 

DISCUSSION

The main findings of our study are as follows: a) ELCA was associated with a high rate of technical success in severely calcified coronary lesions, whether isolated or combined with other plaque modification techniques, with an acceptable ELCA-related complications rate. b) The success rate was higher in undilatable than in uncrossable lesions and was 100% in peri-stent lesions (stent underexpansion or ISR).

As described in previous series, calcified lesions are associated with higher rates of PCI failure, complications, morbidity, and mortality.2,16 Although ELCA is known to have no direct effect on calcium, calcified atheromatous plaques have a mixed composition, including lipids, collagen, and other protein fibers.1,17 The interaction of ELCA with these components, due to its photochemical, photothermal and photokinetic properties, modifies the plaque structure, thus facilitating angioplasty in lesions with severe calcification.17 Moreover, in some cases, as occurs in our series, ELCA is complementary to other plaque modification techniques, allowing the passage of the microcatheter to introduce specific atherectomy guidewires, or even to allow the passage of the lithotripsy balloon.14,15 The RASER technique was used in 2 patients and the ELCA-tripsy technique in another patient with technical success in all 3 of them.

There is a lack of contemporary specific series on the use of ELCA in lesions with severe calcification, and data available in the medical literature are contradictory. Bilodeau et al.18 reported high procedural (93%) and clinical (86%) success in a series of 95 patients with complex coronary lesions, of which 57 had significant calcification. The Laser Veterans Affairs (LAVA) Multicenter Registry7 evaluated the use of ELCA in 131 target complex coronary lesions, of which 62% were moderately or severely calcified lesions, globally reporting 90% technical and 88.8% procedural success rate, which is consistent with our results. In the LEONARDO study,19 in which 75% of lesions were calcified, high laser energy levels were shown to be safe and effective (success rate 93.7%). In our series, the highest fluence and frequency were required in 60% of the procedures, with a similar success rate.

Nowadays, the main indication of ELCA is treatment of uncrossable and undilatable lesions. In uncrossable lesions, the laser catheter can be advanced over any 0.014¨angioplasty guidewire that crosses the lesion, unlike other plaque modification techniques. In a multicenter US registry, the success rate for laser-assisted PCI in uncrossable balloon CTO was 95%, which was higher than that for RA (89%) in this setting.20 In a retrospective study by Karacsonyi et al.,21 laser use in balloon-uncrossable and balloon-undilatable CTO was associated with higher technical (91.5% vs 83.1%) and procedural (88.9% vs 81.6%) success rates compared with cases without the use of laser. Ojeda et al.9 conducted a multicenter registry of 126 uncrossable lesions and reported ELCA success of 81.8%. In that registry, severe calcification was independently associated with ELCA failure, a finding already described in a previous study.22 In our series (with severe calcification in 100% of patients), the overall ELCA success rate was 91.25%, but the ELCA success in uncrossable lesions was lower than in undilatable lesions (78.1% vs 100%) and similar to that in the series by Ojeda et al.9 The lower success rate in uncrossable and severely calcified lesions can probably be explained by the different plaque composition and calcium distribution. Furthermore, the higher rate of use of intracoronary imaging could also be associated with better results (72.5% in our series compared with 22.5% reported by Ojeda et al.9). Of note, an ELCA success of 78.1% in uncrossable lesions with severe calcification could be a reasonable result, considering that, if even a microcatheter cannot cross the lesion, ELCA may be the only alternative for revascularization.

In other scenarios, the ELCA success rate of our series was high and similar to that of other series. An ELCA success rate of 86% to 93% has been reported in CTOs.8,23 RA in CTO has been associated with similar success rates (89%-95.6%)24,25 but with a high rate of slow/no flow phenomena.24 In patients with stent underexpansion and ISR, ELCA is feasible and effective,26,27 with 100% ELCA success in our series.

Intravascular imaging is useful to guide calcified coronary stenosis PCI.28,29 Contemporary rates of intravascular imaging for complex PCI remain low.30 In our study, intracoronary imaging was used in 72 procedures (73.4%), and intracoronary imaging-guided procedures resulted in a higher success rate. Its lower use in uncrossable lesions can probably be explained by the fact that the intravascular ultrasound/OCT catheter cannot cross the lesion, rather than necessarily being the reason for the lower success rate in this setting.

Limitations

Our study has some limitations. First, it is an observational study with a small sample size. However, to the best of our knowledge, our study represents the largest series of ELCA specifically performed in severely calcified lesions in contemporary PCI. Second, the severity of lesion calcification was initially assessed by conventional coronary angiography, which has only low to moderate sensitivity compared with intravascular ultrasound or OCT. In addition, sometimes the calcium observed by conventional angiogra­- phy is adventitious, thus not affecting balloon dilation or stent expansion with conventional techniques. However, the use of intracoronary imaging techniques was higher than in previous series and confirmed the severity of calcification in all patients. In addition, a significant number of cases consisted of uncrossable lesions, limiting the use of intracoronary imaging to define the calcification from the beginning of the procedure. Finally, the operators involved in this study were experienced ELCA operators. This may limit the generalizability of our results since ELCA is not available in most centers and requires a learning curve.

CONCLUSIONS

ELCA is a useful tool in severe calcification lesions, with a high success rate, especially in the setting of undilatable or peri-stent lesions. The technique is also reasonably safe, given that it is used in highly complex procedures. Future randomized studies will shed light on its role in the management of severe calcified coronary lesions.

FUNDING

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

ETHICAL CONSIDERATIONS

All patients signed an informed consent form and approval was obtained from the ethics committee of the center. The study has taken into consideration sex and gender variables according to SAGER guidelines.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

No artificial intelligence tool has been used during the preparation of this work.

AUTHORS’ CONTRIBUTIONS

A. Jurado-Román conceived and designed the study. L. Cobarro and A. Jurado-Román performed the analysis and wrote the initial draft. L. Cobarro, A. Jurado-Román, D. Tébar-Márquez, S. Vera-Vera, A. García-Escobar, C. Ugueto, C. Contreras, B. Rivero, S. Jiménez-Valero, G. Galeote, and R. Moreno collected the data and reviewed the final version of the manuscript.

CONFLICTS OF INTEREST

R. Moreno is associate editor of REC: Interventional Cardiology; the editorial procedure established in the journal has been followed to ensure impartial handling of the manuscript.

A. Jurado-Román is proctor of Philips-Biomenco, Boston Scientific, CSI-World Medica and Medtronic Inc and has received speaker fees from Boston Scientific, Abbott Vascular, World Medica, Biotronik, Philips-Biomenco, and Inari. R. Moreno has received speaker fees from Medtronic Inc, Boston Scientific, Abbott vascular, Biosensors, Biotronik, Edwards Lifesciences, AMGEN, AstraZeneca, and Daiichi Sankyo New Vascular Therapies and Biosensors.

REFERENCES

INTRODUCTION

Contrast induced-acute kidney injury (CI-AKI) is a dreaded complication after diagnostic and interventional angiographic procedures and is linked to increased morbidity and mortality. In a large recent meta-analysis, the pooled incidence of CI-AKI after coronary angio­graphy was 12.8%, with 95% confidence interval (95%CI) 11.7%-13.9%, and the associated mortality was 20.2% (95%CI, 10.7%-29.7%).1 Multiple risk factors have been identified: contrast medium volume (CMV), advanced age (> 75 years), diabetes, anemia, conditions associated with hypotension, and ejection fraction (EF) < 40%.2,3 Many of these risk factors are included in the Mehran score,2 which identifies 4 risk classes of contrast-induced nephropathy (CIN) after PCI: low (≤ 5 points), moderate (6-10 points), high (11-15 points), and very high (≥ 16 points). The Mehran score and the recent Mehran 2 score4 assign 1 point for each 100 mL of CMV up to a dose of 299 mL. Because volume depletion increases the CM concentration in renal tubules, the main preprocedural measure to reduce the occurrence of CI-AKI is intravenous administration of normal saline before and after the procedure, because other solutions provide no benefits5; hydration should be started 12 hours before and continued for 24 hours after the procedure at 1 mL/kg/h or 0.5 mL/kg/h if EF < 35% or New York Heart Association (NYHA) class > 2.6 Another means of decreasing CM concentration in the kidneys is the DyeVert Contrast Reduction system (Osprey Medical Inc, United States), which reduces the amount of CMV delivered to patients during angiographic procedures, with noninferior image quality as attested by independent reviewers.7,8 The DyeVert, DyeVert Plus and DyeVert Plus EZ are used in conjunction with manual contrast injection, and the DyeVert Power XT is used with automated contrast injection; the latter system has been little studied. The main aim of our study was to evaluate the effectiveness of the DyeVert Power XT system in reducing CM delivery during PCI.

METHODS

Study population

This single center, observational study was performed in patients who underwent PCI between September 2020 and December 2022 with the DyeVert Power XT system (case group) and in patients who underwent PCI during a similar period without the use of the device (control group).

Inclusion criteria for both groups were as follows: chronic kidney disease (CKD) [estimated glomerular filtration rate (eGFR) < 60 mL/min/m²] and/or need for a complex PCI with the likelihood of receiving a large amount of CM; previous coronary artery bypass graft (CABG); chronic total occlusion (CTO) (complete blockage of a coronary artery lasting at least 3 months); bifurcation; and left main and/or multivessel disease (at least 2 vessels involved).

The exclusion criterion for both groups was the presence of end-stage kidney failure on dialysis treatment. We collected laboratory, instrumental, clinical, and procedural variables in the case and control groups. Definitions of all these variables are reported in table 1, table 2, table 3, and table 4. For the variables included in the Mehran score, we used the same descriptions as those used in the score. eGFR was calculated by the Modification of Diet in Renal Disease (MDRD) 4-variable equation, left ventricular EF by 2-dimensional echocardiography during hospitalization and before arrival in the catheterization laboratory, and the risk of any post-PCI CIN by the Mehran score. Bifurcation/left main treatment (with single/double stent) consisted of the proximal optimization technique (POT) with kissing balloon inflation and eventually re-POT in all cases. Total CMV represents the volume that would have been delivered if DyeVert had not been used, ie, the sum of CMV delivered to patients and the CMV saved by DyeVert. CM injection flow was 4 and 3 mL/sec for the left and right coronary artery, respectively.

 

Table 1. Laboratory, instrumental, clinical characteristics, and Mehran score in the overall population and according to incidence of CI-AKI in the case group

Characteristics	Overall population (n = 101)	No CI-AKI (n = 87)	CI-AKI (n = 14)	P
Laboratory and istrumental characteristics
eGFR, mL/min	51 ± 18	52 ± 19	45 ± 16	.18
HCT	38.6 ± 4.9	39.1 ± 4.8	35.5 ± 4.8	.01*
EF	50 [35-55]	50 [40-55]	30 [28-36]	< .001*
CKD [eGFR < 60 mL/min/ 1.73 m2]	73 (72.3)	63 (72.4)	10 (71.4)	1
Anemia [male HCT < 39%, female HCT < 36%]	48 (47.5)	38 (43.7)	10 (71.4)	.10
Clinical characteristics
Age, years	74 (68-80)	73 (67-80)	75 (74-81)	.09
Age > 75 years	39 (38.6)	32 (36.8)	7 (50)	.52
Male sex	80 (79.2)	68 (78.2)	12 (85.7)	.73
Overweight [body mass index ≥ 25]	52 (51.5)	46 (52.9)	6 (42.9)	.68
Hypertension	78 (77.2)	70 (80.5)	8 (57.1)	.08
Diabetes	48 (47.5)	40 (46)	8 (57.1)	.62
Dyslipidemia	68 (67.3)	57 (66)	11 (79)	.51
Current smoker	24 (23.8)	20 (23)	4 (28.6)	.74
Former smoker	35 (34.7)	32 (36.8)	3 (21.4)	.37
CHF [NYHA class ≥ 3 and/or history of pulmonary edema]	37 (36.6)	25 (28.7)	12 (85.7)	< .001*
Acute coronary syndrome presentation	38 (37.6)	31 (35.6)	7 (50)	.46
Hypotension [Systolic arterial pressure < 80 mmHg for ≥ 1 h requiring inotrope]	4 (4)	2 (2.3)	2 (14.3)	.09
Mehran score
Mehran CI-AKI risk class:
Low	24 (23.8)	24 (27.6)	0 (0)	.04*
Moderate	26 (25.7)	24 (27.6)	2 (14.3)	.51
High	34 (33.7)	29 (33.3)	5 (35.7)	1
Very high	17 (16.8)	10 (11.5)	7 (50)	.002*
Mehran score, points	11 ± 5	10 ± 5	15 ± 4	< .001*
Values are expressed as No. (%), mean ± standard deviation, or median [first quartile-third quartile]. *Statistically significant P-value (P < .05). CHF, congestive heart failure; CI-AKI, contrast induced-acute kidney injury; CKD, chronic kidney disease; EF, ejection fraction; eGFR, estimated glomerular filtration rate; HCT, hematocrit; NYHA, New York Heart Association.

 

Table 2. Procedural characteristics in the overall population and according to incidence of CI-AKI in the case group

Characteristics	Overall population (n = 101)	No CI-AKI (n = 87)	CI-AKI (n = 14)	P
Procedural characteristics (angiography/PCI complexity/complications)
Previous CABG	20 (19.8)	18 (20.7)	2 (14.3)	.73
CTO [complete blockage of a coronary artery lasting at least 3 months]	12 (11.9)	11 (12.6)	1 (7.1)	1
No. vessels treated in the same procedure:
1	57 (56.4)	52 (59.8)	5 (35.7)	.09
2	40 (39.6)	32 (36.8)	8 (57.1)	.15
3	4 (4)	3 (3.4)	1 (7.1)	.45
No. bifurcations treated in the same procedure:
0	67 (66.3)	58 (66.7)	9 (64.3)	1
1	31 (30.7)	27 (31)	4 (28.6)	1
2	3 (3)	2 (2.3)	1 (7.1)	.36
Left main treatment	25 (24.8)	20 (23)	5 (35.7)	.33
Stent, number	2 [1-3]	2 [1-3]	2 [1-3]	.75
Stent lenght, mm	52 [31-88]	51 [30-91]	57 [36-73]	.95
Perforation	3 (3)	3 (3.4)	0 (0)	1
IABP use	1 (1)	0 (0)	1 (7.1)	.14
Rotablator use	3 (3)	1 (1.1)	2 (14.3)	.05
Procedural characteristics (others)
Radial access	88 (87.1)	75 (86.2)	13 (92.9)	.69
Femoral access	27 (26.7)	21 (24.1)	6 (42.9)	.19
Operator
L	52 (51.5)	47 (54)	5 (35.7)	.20
A	30 (29.7)	26 (29.9)	4 (28.6)	1
B	4 (4)	3 (3.4)	1 (7.1)	.46
V	13 (12.9)	10 (11.5)	3 (21.4)	.38
S	2 (1.9)	1 (1.1)	1 (7.1)	.26
Contrast medium type:
Iomeprol 350	9 (8.9)	7 (8)	2 (14.3)	.61
Iohexol 350	13 (12.9)	11 (12.6)	2 (14.3)	1
Iodixanol 320	79 (78.2)	69 (79.3)	10 (71.4)	.50
Contrast medium dose delivered, mL	242 [189-300]	240 [188-306]	258 [195-277]	.95
Total contrast medium dose [delivered plus saved], mL	355 ± 110	354 ± 79	356 ± 106	.95
IVUS use	24 (23.8)	23 (26.4)	1 (7.1)	.18
Data are expressed as No. (%), mean ± standard deviation, or median [first quartile-third quartile]. CABG, coronary artery bypass graft; CI-AKI, contrast induced-acute kidney injury; CTO, chronic total occlusion; IABP, intra-aortic balloon pump; IVUS, intravascular ultrasound.

 

Table 3. Laboratory, instrumental, clinical characteristics, and Mehran score of cases and controls in the matched group

Characteristics	No DyeVert (n = 49)	DyeVert (n = 49)	P	Standardized mean difference
Laboratory and istrumental characteristics
eGFR, mL/min	53 ± 18	51 ± 18	.70	0.11
HCT	37.8 ± 4.1	38.2 ± 4.9	.68	0.08
EF	50 [40-55]	50 [35-55]	.68	0.13
CKD [eGFR < 60 mL/min/1.73 m2]	36 (73.5)	34 (69.4)	.82	0.09
Anemia [male HCT < 39, Female HCT < 36]	27 (55.1)	24 (49)	.69	0.12
Clinical characteristics
Age, years	75 ± 9	75 ± 9	.96	0.01
Age > 75 years	26 (53.1)	24 (49)	.84	0.08
Male sex	38 (77.6)	41 (83.7)	.61	0.15
Overweight [body mass index ≥ 25]	22 (44.9)	24 (49)	.84	0.08
Hypertension	33 (67.3)	37 (75.5)	.50	0.19
Diabetes	19 (38.8)	20 (40.8)	1	0.04
Dyslipidemia	28 (57.1)	32 (65.3)	.53	0.17
Current smoker	11 (22.4)	10 (20.4)	1	0.05
Former smoker	16 (32.7)	18 (36.7)	.83	0.09
CHF [NYHA class ≥ 3 and/or history of pulmonary edema]	15 (30.6)	15 (30.6)	1	< 0.01
Acute coronary syndrome presentation	27 (55.1)	25 (51)	.84	0.08
Hypotension [systolic pressure < 80 mmHg for ≥ 1 h requiring inotrope]	2 (4.1)	2 (4.1)	1	< 0.01
Mehran score
Mehran CI-AKI risk class:
Low	12 (24.5)	10 (20.4)	.63	0.09
Moderate	12 (24.5)	15 (30.6)	.50	0.14
High	13 (26.5)	15 (30.6)	.65	0.09
Very high	12 (24.5)	9 (18.4)	.46	0.16
Mehran score, points	11 ± 6	11 ± 6	.86	0.04
Data are expressed as No. (%), mean ± standard deviation, or median [first quartile-third quartile]. CHF, congestive heart failure; CI-AKI, contrast induced-acute kidney injury; CKD, chronic kidney disease; EF, ejection fraction; eGFR, estimated glomerular filtration rate; HCT, hematocrit; NYHA, New York Heart Association.

 

Table 4. Procedural characteristics of cases and controls in the matched group

Characteristics	No DyeVert (n = 49)	DyeVert (n = 49)	P	Standardized mean difference
Procedural characteristics (angiography/PCI complexity/complications)
Previous CABG	8 (16.3)	6 (12.2)	.56	0.10
CTO [complete blockage of a coronary artery lasting at least 3 months]	6 (12.2)	8 (16.3)	.77	0.13
No. vessels treated in the same procedure:
1	32 (65.3)	29 (59.2)	.53	0.12
2	15 (30.6)	17 (34.7)	.67	0.08
3	2 (4.1)	3 (6.1)	1	0.10
No. bifurcations treated in the same procedure:
0	33 (67.3)	31 (63.3)	.67	0.09
1	15 (30.6)	16 (32.7)	.83	0.04
2	1 (2.1)	2 (4)	1	0.12
Left main treatment	12 (24.5)	13 (26.5)	1	0.05
Stent, number	2 [1-3]	2 [1-3]	.30	0.15
Stent lenght, mm	46 [30-85]	52 [33-97]	.41	0.13
Perforation	2 (4.1)	1 (2)	1	0.12
IABP use	0 (0)	1 (2)	1	0.20
Rotablator use	0 (0)	2 (4.1)	.49	0.24
Procedural characteristics (others)
Radial access	41 (83.7)	45 (91.8)	.35	0.24
Femoral access	11 (22.4)	15 (30.6)	.49	0.18
Operator:
L	24 (49)	24 (49)	1	< 0.01
A	17 (34.7)	17 (34.7)	1	< 0.01
B	5 (10.2)	3 (6.1)	.71	0.20
V	3 (6.1)	5 (10.2)	.71	0.12
Contrast medium type:
Iomeprol 350	7 (14.3)	4 (8.2)	.34	0.21
Iohexol 350	9 (18.4)	10 (20.4)	.80	0.06
Iodixanol 320	33 (67.3)	35 (71.4)	.66	0.10
IVUS use	10 (20.4)	11 (22.4)	1	0.05
Data are expressed as No. (%), mean ± standard deviation, or median [first quartile-third quartile]. CABG, coronary artery bypass graft; CI-AKI, contrast induced-acute kidney injury; CTO, chronic total occlusion; IABP, intra-aortic balloon pump; IVUS, intravascular ultrasound.

Image quality was evaluated by operators during the procedures. When quality was inadequate, exclusion of the device from the CM line was allowed for the shortest possible time.

AKI was defined as a rise in the concentration of serum creatinine ≥ 0.3 mg/dL within 48 hours after CM administration from the baseline value obtained before CM injection; further measurements after 48 hours were collected in patients with worsening kidney function; for its prevention, all patients received hydration with sodium chloride 0.9% intravenous solution at a rate of 1 or 0.5 mL/kg/h, as appropriate. The severity of AKI was defined according to Kidney Disease Improving Global Outcome (KDIGO) stages.

The research reported was performed in accordance with recommendations for clinical investigation (Declaration of Helsinki of the World Medical Association, October 2013) and was approved by an ethics committee. We declare that relevant informed consent was obtained from all participants and is available.

Objectives

In the case group, we evaluated the following: a) the amount of CMV saved using DyeVert and image quality; b) the rate and severity of CI-AKI and the rate of in-hospital all-cause death; c) laboratory, instrumental, clinical, and procedural differences in the 2 subgroups defined on the basis of the incidence of AKI; and d) independent predictors of CI-AKI.

In the overall population of the case and control groups, we performed propensity score matching (PSM) to obtain a group of patients with a sufficiently good balance (matched group), in which we evaluated the following: a) differences in CMV, and b) rate and severity of CI-AKI.

Statistical analysis

Categorical variables are expressed as the number and percentage of patients. Continuous parametric data are reported as the mean ± standard deviation and continuous nonparametric data as the median [lower and upper quartile]; for assessment of normality, the Kolmogorov test was used. Patients’ categorical variables were compared using the chi-squared test (with Yates’ correction for continuity in the case of variables with only 2 categories) or the Fisher exact test, as appropriate. The unpaired t-test was used for continuous parametric variables and the Mann-Whitney U-test for continuous nonparametric variables; the same tests were used in the matched group. On univariate analysis, significance was defined as P < .05. To establish the independent predictors of AKI, we performed multivariable logistic regression analysis. Variables were selected according to significance in the univariate analysis. The chosen method was stepwise backward regression with a maximum of 20 iterations. Multicollinearity was assessed with tolerance and variance inflation factor (VIF) values. Receiver operating characteristic (ROC) curves were used to establish the optimal cutoffs of independent predictors for the diagnosis of AKI. To perform PSM, the algorithm used was nearest neighbor matching 1:1 with a caliper size of ± 0.2. Statistical analyses were performed using SPSS for Windows, release 29, with R 4.2 implementation to perform PSM.

RESULTS

Analysis in the case group

A total of 101 patients (median age 74 [68-80] years, male sex 79.2%, CKD 72.3%) underwent PCI with the use of the DyeVert Power XT system.

In the overall population of the case group, mean hematocrit (HCT) was 38.6 ± 4.9 %, median EF was 50% [35%-55%], and mean Mehran score was 11 ± 5 points.

Congestive heart failure (CHF) was present in 37 patients (36.6%), Mehran CI-AKI very high-risk class was present in 17 patients (16.8%) and Mehran CI-AKI low-risk class was present in 24 patients (23.8%) (table 1).

We enrolled 20 patients (19.8%) with previous CABG, 12 (11.9%) with CTO, 34 (33.7%) with bifurcations, 25 (24.8%) with left main coronary artery disease, and 44 (43.6%) with multivessel disease. Delivered CM was 242 (189-300) mL, total CM was 355 ± 110 mL, and saved CM was 114 ± 42 mL, with an average of 32% of the total CMV (table 2). In almost all patients (n = 96, 95% of patients), image quality was adequate, while the device was excluded to make it adequate for the shortest possible time in 5 patients. Without these exclusions, saved CMV would have been slightly higher and with trivial changes with regard to the comparison with controls: 33% of the total, a value derived from patients without exclusions (n = 96).

A total of 14 (13.9%) patients developed CI-AKI (AKI-KDIGO 1, 2, 3: 6.9%, 3%, and 4%, respectively). The results of the univariate analysis for the overall population and according to the incidence of CI-AKI in the case group are reported in table 1, table 2, and figure 1.

Figure 1. In-hospital all-cause mortality rate according to onset of CI-AKI in the case group. CI-AKI, contrast induced-acute kidney injury.

 

Compared with patients not developing CI-AKI, those in the CI-AKI subgroup had lower HCT values (35.5 ± 4.8 vs 39.1 ± 4.8; P = .01), lower EF values (30 [28-36] vs 50 [40-55]; P < .001) and higher Mehran score values (15 ± 4 vs 10 ± 5; P < .001).

In addition, the first patients more frequently had CHF [12 (85.7%) vs 25 (28.7%); P <.001] and Mehran CI-AKI very high-risk class (7 [50%] vs 10 [11.5%]; P =.002) and less frequently had Mehran CI-AKI low-risk class (0 [0%] vs 24 [27.6%]; P = .04).

No significant differences were found in the remaining laboratory, instrumental, or clinical features or the procedural variables between the 2 subgroups; in particular, CM was slightly higher in CI-AKI patients: 258 [195-277] vs 240 [188-306] mL, total 356 ± 106 vs 354 ± 79 mL; P = .95 for both variables delivered. In the multivariate analyses, independent predictors of CI-AKI were HCT (OR, 0.86, 95%CI, 0.74-0.99; P = .04) and EF (OR, 0.88, 95%CI, 0.82-0.95; P = .001); the percentage accuracy in classification of the model was 88%, while tolerance and VIF values (0.99 and 1.01, respectively) showed no multicollinearity. The HCT ROC curve showed the following values: area under curve (AUC) 0.71 with 95%CI 0.56-0.87; P = .01; a cutoff of 36.3% had the best sensitivity (72%) and specificity (71%) for the outcome (figure 2). The EF ROC curve showed the following values: AUC 0.83 with 95%CI 0.72-0.94; P = .001; a cutoff of 37% had the best sensitivity (82%) and specificity (79%) (figure 2); therefore, our best predictor was EF < 40%.

Figure 2. Receiver operating characteristic curves showing the diagnostic ability of HCT and EF for the diagnosis of CI-AKI in the case group. CI-AKI, contrast induced-acute kidney injury; EF, ejection fraction; HCT, hematocrit.

 

There were 4 in-hospital all-cause deaths overall, 2 deaths in each subgroup (CI-AKI and no–CI-AKI subgroups), as shown in figure 1.

Analysis in the matched group

After the matching process, 49 patients remained in the control (no DyeVert) and case (DyeVert) groups with no significant imbalance (ie, standardized mean differences < ± 0.25), as reported in table 3 and table 4. As shown in figure 3, delivered CM was slightly lower in the DyeVert group than in the no-DyeVert group, with no significant difference (252 ± 80 vs 267 ± 101 mL; P = .42), while total CM was significantly higher in the DyeVert group (354 ± 110 vs 267 ± 101 mL; P < .001). The CI-AKI rate was slightly lower in the case group than in the control group (14.3% vs 16.3%; P = .99) with slightly more advanced stages of AKI in controls (table 1 of the supplementary data), without significance.

Figure 3. Contrast medium in cases and controls in the matched group. CM, contrast medium; blue dots represent the median values; vertical black lines represent the standard deviations.

 

DISCUSSION

In the case group, the DyeVert Power XT system saved 32% of CM and image quality was adequate in almost all cases; the only independent predictors of CI-AKI were HCT and EF.

In the matched group, total CM was higher in cases than in controls. After diversion by the device, delivered CM was slightly lower in cases than in controls, but without significance. The reduction in CI-AKI was also nonsignificant.

The DyeVert system is a second-generation device to reduce the amount of CM delivered to patients during angiographic procedures. The first generation was the AVERT system (Osprey Medical Inc), which showed a relative reduction of approximately 23% in CMV among PCI patients compared with controls; the use of the device did not reduce the AKI rate.9 DyeVert Power XT is used in combination with automatic injection; few data are available in this context, being limited to 2 studies that investigated 2610 and 9 patients,11 without a control group. There are more data on manual injection (1696 patients, 15 studies); all these 17 studies were collectively analyzed in the meta-analysis by Tarantini et al.12

In that meta-analysis, the mean saved CMV in the DyeVert group was reported by 7 observational studies and ranged from 34% to 47% of total CMV; the pooled estimate value was approximately 39.5% using manual CM injection systems; of note, the lowest value (34%) was achieved using DyeVert Power XT. We found a similar value in the DyeVert (case) group. These reduced values compared with manual systems may be related to different pressures generated during automatic contrast injection.

In our case group analysis, CMV was not significantly correlated with the occurrence of CI-AKI, which instead was independently predicted by lower HCT and EF values, which are known risk factors, as shown by Mehran scores.2,4 EF was also an independent predictor in the study by Briguori et al.13 Our findings confirm the importance of first identifying the variables (eg, those in the Mehran or Mehran 2 scores)2,4 that classify patients at higher risk of CI-AKI to apply appropriate preventive strategies. In the present study, these patients were identified by HCT and EF and consequently the latter variables (especially EF) may be more important predictors than CMV, which is normally used during PCI in the general population. In the above-mentioned scores, CMV was also an independent predictor of CI-AKI and, consequently, using the smallest possible value of CMV is still important, especially in higher risk patients. DyeVert thus has the potential benefit of reducing CIN, depending on its efficacy compared with controls, which was evaluated in the above-mentioned meta-analysis and in the present study.

In the meta-analysis, approximately half of the studies included controls for comparison. Delivered CM was usually lower in DyeVert patients than in controls. In these cases, the difference ranged from 22 to 50 mL,12 with the highest differences being reported in the studies by Tajti et al. (200 [153-256] vs 250 [170-303] mL; P = .04) and Briguori et al. (99 ± 50 vs 130 ± 50 mL; P <.001).13,14 Delivered CMV was slightly higher (difference of 2 mL) in the DyeVert group only in the study by Bunney et al.15 The pooled analysis showed a significant decrease in delivered CMV with DyeVert use relative to the control group. Of note, details about prior CABG, CTO and left main treatment were reported only in 1 work14 and the number of vessels treated was reported only in another work.13 The treatment of bifurcations and differences in operators were not reported. All these procedural characteristics, which may influence the amount of CM delivered during PCI, were included in our study and we used a matched group with a sufficiently good balance in the studied characteristics.

In our matched group, delivered CM was lower in the case group than in the control group, but the difference was slight and nonsignificant, while total CM (also called attempted in the meta-analysis) was significantly higher in the case group than in the control group. Consequently, the net practical benefit of the device in terms of spared CM was low. In our work, procedural characteristics (eg, procedural complexity), which could cause discrepancies in CM injections, were balanced in the matched group. Based on these findings, we believe that the control group required more prolonged and/or a greater number of contrast injections (and consequently more total CM) to achieve adequate image quality. In previous studies, adequate image quality was achieved with DyeVert in 98% of cases,12 a value similar to ours, but those studies did not discuss the need for prolonged injections and more total CM compared with controls to maintain image quality when DyeVert is used. Few data are available on total CM, but previous studies reporting this information indicate that total CM was higher in DyeVert patients than in controls (Briguori et al., P-value almost significant; Kutschman et al., P-value not reported).12,13,16

The reduction in CI-AKI in the present study was not statistically significant. In the meta-analysis, the pooled relative risk for CI-AKI associated with DyeVert system use was 0.60 (95%CI, 0.40-0.90; P = .01), which was a result derived from 5 studies. Moreover, in a recent abstract not included in the meta-analysis, postprocedure eGFR values among patients undergoing coronary and/or peripheral angiography were significantly more stable in the DyeVert group than in controls.17

Analysis of the 5 above-mentioned studies separately revealed that our results are mainly in agreement; indeed, the relative risk was significantly lower in only 1 study in the nonpooled analysis.13

The type of CM was not associated with the occurrence of CI-AKI; as recommended,18 we used iso-osmolar (Iodixanol 320) or low-osmolar (Iomeprol 350 or Iohexol 350) contrast agents to prevent CIN. Given the presence of more favorable evidence,19 we preferred to use the iso-osmolar agent and reserved the other agents to low-risk patients.

Study limitations

Our study has some limitations. First, the sample size was relatively small. Second, the study design was single center, observational and retrospective, although we performed PSM to reduce potential confounding bias. Third, we excluded patients not meeting the inclusion criteria, as they were usually at low risk of CI-AKI. Therefore, our results should be generalized with caution, since the analyzed patients may be not representative of the general population. In this work the variable of sex has not been taken into account in accordance with the SAGER guidelines.

CONCLUSIONS

The DyeVert Power XT system saved 32% of CM, but only HCT and EF were independent predictors of CI-AKI and the main predictor was EF < 40%. Therefore, these variables (especially EF) may be more important than CMV, which is normally used during PCI in the general population.

PCI with this system required more total CM compared with that in controls to achieve adequate image quality. Consequently, after CM saving by the device, delivered CM was only slightly lower than CM in controls (mean difference of 15 mL) and this difference was nonsignificant. Therefore, the net practical benefit of the system was low. Equally, the reduction in CI-AKI (14.3% vs 16.3%) was not statistically significant.

Future studies are needed to confirm these results.

FUNDING

The authors did not receive any grants for this research.

ETHICAL CONSIDERATIONS

The work has been approved by an Ethics Committee/institution. Informed consent of patients was obtained and archived for the publication of their cases. In this work the variable of sex has not been taken into account in accordance with the SAGER guidelines.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

We didn’t use artificial intelligence for the development of our work.

AUTHORS’ CONTRIBUTIONS

F. Vergni, M.Arioti, and M.Leoncini contributed to the design of the work. F. Vergni, M. Arioti, V. Boasi, F.A. Sánchez, M.Leoncini, and F. Ferrari contributed to the acquisition of data. F. Vergni analyzed the data. F. Vergni, M.Arioti, V. Boasi, F.A. Sánchez, M. Leoncini, and F. Ferrari contributed to the interpretation of the data. F. Vergni and M.Arioti contributed to the drafting of the work. F. Vergni, M. Arioti, V. Boasi, F.A. Sánchez, M.Leoncini, and F. Ferrari revised the work and approved the final version to be published.

CONFLICTS OF INTEREST

The authors have no conflicts of interest to declare.

 

REFERENCES

INTRODUCTION

Over the past few years, the term INOCA (ischemia with nonobstructive coronary arteries) has established to define patients with signs or symptoms of ischemic heart disease without angiographically significant obstructive coronary artery disease.1 In these patients, coronary microvascular or epicardial vessel dysfunction could be the pathophysiological mechanism triggering the symptoms and ischemic impairment.2

Currently, the invasive study of microvascular function in patients with INOCA is a recommendation IIa according to the clinical practice guidelines of the European Society of Cardiology.3 What it does is measure the parameters that show its functional or structural status like coronary flow reserve (CFR) or index of microcirculatory resistance (IMR).4

Recently, the possibility of measuring absolute coronary blood flow (Q) and microvascular resistance (R) by continuous thermodilution with the infusion of a physiological saline solution through a specific coronary microcatheter has been described. This technique has potential advantages like its independence from the operator or not needing pharmacologically induced hyperemia.5

The objective of this study is to estimate the prevalence of the different endotypes of patients with INOCA and analyze the correlation between the measurements obtained by continuous thermodilution and the traditional method of intracoronary boluses of physiological saline solutions.

METHODS

This was a prospective and consecutive study of 60 referred patients due to symptoms or signs of myocardial ischemia without angiographically significant coronary artery stenosis on the visual estimate (< 50%) or after functional assessment (resting full-cycle ratio [RFR] > 0.89 or fractional flow reserve [FFR] > 0.80). Severe valvular heart disease, acute coronary syndrome, decompensated heart failure, and any clinical or anatomical condition where the study of microcirculation and vasoreactivity would be considered unnecessary were excluded.

All microcirculation and vasoreactivity studies were scheduled and second-staged. Nitrates and calcium antagonists were withdrawn prior to conducting the tests.

The coronary angiography was performed based on the routine clinical practice via radial access. A spasmolytic cocktail of 200 µg of nitroglycerin was administered. The target artery was the left main coronary artery.

The study was approved by the center ethics research committee and the patients’ written informed consent was obtained.

Vasoreactivity test

First, the vasoreactivity test was performed. Patient monitoring included precordial leads, and baseline angiograms were performed using 2 different projections. The sequential administration of acetylcholine was followed by increasing doses of 2 µg, 20 µg, and 100 µg in intracoronary bolus for 2 min. In the presence of significant bradycardia, the injection was interrupted, and if considered appropriate, it was re-administered at a slower rate. A follow-up angiogram was performed after every dose. In the presence of severe symptoms, changes to the echocardiogram or epicardial spasm 200 µg of intracoronary nitroglycerin were administered.

The test was considered positive based on the criteria established by the COVADIS (Coronary vasomotor disorders international study) group: epicardial spasm in the presence of chest pain, changes to the echocardiogram, and constriction ≥ 90%, and microvascular spasm in the presence of chest pain, and changes to the echocardiogram without epicardial spasm ≥ 90%.6

Indices obtained with continuous thermodilution

After the administration of unfractionated heparin (70 IU/kg), a pressure-temperature sensor guidewire Pressure Wire X (Abbott, United States) was inserted and pressures at the catheter distal border were equalized. The guidewire was advanced until it reached the left anterior descending coronary artery distal segment.

Resting full-cycle ratio was registered to confirm the lack of hemodynamically significant epicardial stenoses (RFR > 0.89).

Afterwards, a specific Rayflow (Hexacath, France) microcatheter for intracoronary infusion was placed in the left anterior descending coronary artery proximal segment. After confirmation that the guidewire sensor was, at least, 3 cm distal to the tip of the microcatheter, the intracoronary infusion of a physiological saline solution at room temperature and at a dose of 20 mL/min was started using an injector pump to induce hyperemia.

Pressure-temperatures curves were registered using Coroventis software (Abbott, United States). When the distal temperature drop was stabilized, the sensor was withdrawn up to the tip of the microcatheter to determine the infusion temperature.

Afterwards, the injection of the physiological saline solution stopped, and Q (L/min) and R (Wood units) values were obtained automatically (figure 1).

Figure 1. Measurements obtained by continuous thermodilution: DP, distal pressure; Q, absolute coronary blood flow; Q_i, infusion flow (mL/min); R, micro­vascular resistance; T, distal temperature; T_i, infusion temperature.

 

Indices obtained with bolus thermodilution of a physiological saline solution

After completion of the continuous thermodilution study, and once the Rayflow microcatheter was removed, the pressure-temperature guidewire was repositioned in its previous location, and thermodilution curves were registered using the Coroventis software after the vigorous manual injection of 3 intracoronary boluses of 3 mL of a physiological saline solution. Measurements were taken at rest and after inducing hyperemia with a peripheral intravenous bolus of regadenoson (400 µg) resulting in the calculation of CFR and IMR.

Statistical analysis

Continuous variables were expressed as mean ± standard deviation or median [interquartile range]. The categorical ones were expressed as absolute value or percentage. ROC (Receiver operating characteristic) curves were used to estimate the optimal cut-off values for the continuous variables Q and R. The cut-off values established as pathological for CFR < 2 and IMR ≥ 25 were used as the reference framework. Once dichotomized, the variables Q and R were compared to the CFR and IMR values using chi-square tests. One-way ANOVA was used to compare the different quantitative variables. The statistical analysis was performed using the SPSS v 20 statistical software package (IBM, United States).P values < .05 were considered statistically significant.

RESULTS

Study patients

Table 1 shows the baseline characteristics of the 60 patients included in the study. Women (55%) were predominant. Also, there was a high prevalence of cardiovascular risk factors. Most showed typical angina-like clinical signs (76%) and had tested positive to an ischemia test performed before the coronary angiography (60%).

 

Table 1. Clinical and angiographic characteristics (N = 60)

Age (years)	63 ± 10
Women	33 (55%)
Hypertension	39 (65%)
Diabetes	21 (35%)
Dyslipidemia	35 (58%)
Smoking (current or past)	28 (47%)
Previous percutaneous revascularization	4 (7%)
Previous myocardial infarction	3 (5%)
Left ventricular systolic dysfunction	4 (7%)
Ejection fraction (%)	63 ± 8
Clinical presentation
Exertional angina	19 (32%)
Resting angina	13 (22%)
Mixed angina	14 (23%)
Other	14 (24%)
Ischemia test
Ergometry	19 (32%)
Isotopic scintigraphy	18 (30%)
Dobutamine stress echocardiography	3 (5%)
None	20 (33%)
Coronary angiography
Atheromatous disease	22 (37%)
Slow flow	13 (22%)
Data are expressed as no. (%) or mean ± standard deviation.

 

The baseline coronary angiography confirmed that 37% of the patients showed parietal irregularities consistent with atheromatous disease, and 22% had slow coronary flow. The FFR and RFR values were normal in all the cases studied.

Coronary vasoreactivity

As shown on table 2, 60% of the cases (36/60) had a positive response to acetylcholine in the vasoreactivity test. A total of 32% of the cases (19/60) showed severe epicardial vasoconstriction, and 23% (14/60) met the criteria for microvascular spasm. In 3 patients (5%), microvascular spasm was observed concomitantly with the medium dose (20 µg), and epicardial spasm with the high dose (100 µg), which added to the impaired indices of microvascular function was consistent with a mixed endotype.

 

Table 2. Procedural data

Pathological vasoreactivity testing	36 (60%)
Epicardial vasospasm	19 (32%)
Microvascular vasospasm	14 (23%)
Combined vasospasm	3 (5%)
Structural microvascular dysfunction (IMR ≥ 25)	20 (33%)
Isolated	5 (8%)
Associated with epicardial spasm	8 (13%)
Associated with microvascular spasm	4 (7%)
Associated with combined spasm	3 (5%)
CFR < 2	11 (18%)
CFR < 2.5	17 (28%)
RFR	0.93 [0.91-0.94]
FFR	0.90 [0.87-0.93]
Q (mL/min)	170 ([138-219]
R (WU)	496 [381-654]
CFR	3.0 [2.3-4.2]
IMR	20 [12-28]
Data are expressed as no. (%) or median [interquartile range]. CFR, coronary flow reserve; FFR, fractional flow reserve; IMR, index of microvascular resistance; Q, absolute coronary blood flow; R, microvascular resistance; RFR: resting full-cycle ratio.

 

Indices of microvascular function

Both studies—bolus thermodilution and continuous infusion thermodilution—were performed uneventfully in all of the patients. Table 2 shows the values of the measurements of microvascular function obtained with both techniques.

In the continuous infusion study, a median of absolute flow in the left anterior descending coronary artery of 170 mL/min [138-219 mL/min] was described while the median of microvascular resistance was 496 WU [381-654 WU].

A total of 18% of the patients (11/60) had a reduced CFR (CFR < 2) while 33% (20/60) showed elevated resistances (IMR ≥ 25).

The group of patients with microvascular dysfunction due to low CFR with normal IMR (7/60, 12%) with respect to cases with microvascular dysfunction due to high IMR with normal CFR (16/60, 27%) had a clinical profile with a lower mean age (61 ± 11 vs 66 ± 8), and a higher predominance of women (86% vs 58%) although this tendency was not statistically significant.

Table 3 shows the mean transit times (MTT) of bolus thermodilution tests. The cases with low CFR showed significantly shorter baseline MTT (0.48 ± 0.45 vs 1.13 ± 0.70), especially the subgroup of patients with low CFR and high Q (0.31 ± 0.15 vs 0.77 ± 0.68).

 

Table 3. Mean transit times obtained by bolus thermodilution

	Overall (N = 60)	CFR < 2 (N = 11)	CFR < 2 Q > 170 (N = 7)	CFR < 2 Q < 170 (N = 4)
Baseline MTT	1.13 ± 0.70	0.48 ± 0.45*	0.31 ± 0.15*	0.77 ± 0.68
Hyperemic MTT	0.36 ± 0.25	0.35 ± 0.28	0.25 ± 0.14	0.51 ± 0.41
Values (in seconds) are expressed as mean ± standard deviation. CFR, coronary flow reserve; MTT, mean transit time; Q, absolute coronary blood flow. * P < .05 for comparison with the rest of the sample.

 

Figure 2 shows data of coronary flow estimated by MTT measurement divided into 3 groups based on CFR and IMR results. We should mention that patients with low CFR without elevated resistances had significantly high resting flows and hyperemic flows without significant differences compared to the rest while in patients with low CFR and elevated resistances, the opposite phenomenon was described.

Figure 2. Baseline and hyperemic mean flow estimated based on the MTT (1/MTT) and grouped based on the CFR and IMR results. Values are expressed as s^–1. CFR, coronary flow reserve; IMR, index of microvascular resistance; MTT, mean transit time.
* P < .05.

 

Endotypes

Figure 3A shows the percentages of endotypes based on the result of the acetylcholine test and the measurements of CFR and IMR. The most common pattern was microvascular dysfunction (24/60, 40%) followed by the normal study (14/60, 23%). In 20% of the patients (12/60), microvascular dysfunction overlapped with epicardial vasospasm while in 17% of the patients (10/60) isolated epicardial vasospasms were seen.

Figure 3. A: endotype-based classification. Values are expressed as absolute number and percentage.B: mean values of absolute flow and microvascular resistance grouped by endotypes. Q, absolute coronary blood flow; R, microvascular resistance.
* P < .05 with respect to normal study.

 

Table 4 shows how the mechanisms of vasomotor and microvascular dysfunction overlap in many cases.

 

Table 4.Results of the acetylcholine test and bolus thermodilution study (N = 60)

Epicardial spasm	Microvascular spasm	IMR ≥ 25	CFR < 2	Endotype	Cases
−	−	−	−	Normal	14 (23.3%)
+	−	−	−	Epicardial vasospasm	10 (16.7%)
−	+	−	−	Microvascular dysfunction	9 (15.0%)
−	−	+	−	Microvascular dysfunction	5 (8.3%)
−	−	−	+	Microvascular dysfunction	5 (8.3%)
−	+	+	−	Microvascular dysfunction	3 (5.0%)
−	+	−	+	Microvascular dysfunction	1 (1.6%)
−	+	+	+	Microvascular dysfunction	1 (1.6%)
+	−	+	−	Mixed disorder	6 (10.0%)
+	+	+	−	Mixed disorder	2 (3.3%)
+	−	+	+	Mixed disorder	2 (3.3%)
+	−	−	+	Mixed disorder	1 (1.6%)
+	+	+	+	Mixed disorder	1 (1.6%)
Data are expressed as no. (%) CFR, coronary flow reserve; IMR, index of microvascular resistance.

 

The association between epicardial vasospasm and structural microvascular dysfunction (IMR ≥ 25) was the most prevalent combination in cases of mixed disorder (11/12). In turn, this endotype, in continuous thermodilution measurements, showed significant differences compared to the normal pattern, with reduced absolute flow values and elevated resistances (figure 3B) indicative of more serious structural and functional damage.

Concordance among the different indices of microvascular function

The ROC curve analysis of absolute coronary blood flow (Q) with respect to CFR < 2 determined an optimal cut-off value of 170 mL/min (a 64% sensitivity, and a 52% specificity) with an area underthe curve of 0.50 (95% confidence interval [95%CI], 0.33-0.66;P = .97), therefore showing no diagnostic utility.

Given the recent proposal to consider the cut-off value of CFR < 2.57,7 the analysis was performed using this threshold as the reference. In addition, no significant concordance was seen (area under the curve of 0.45 [95%CI, 0.30-0.61;P = .56]).

Regarding R with respect to IMR, an area under the curve of 0.67 (95%CI, 0.51-0.82;P = .04) was obtained, which was indicative of a weak yet significant diagnostic concordance (figure 4). The estimated optimal cut-off value was 435 WU, which was consistent with an 81% sensitivity and a 57% specificity. A total of 66% of cases with IMR ≥ 25 were categorized correctly using this index.

Figure 4. Analysis of the R cut-off value > 435 WU to predict IMR ≥ 25, and scatter plot showing the correlation between IMR and R. IMR, index of microvascular resistance; R, microvascular resistance.

 

The absence of an association between Q and CFR was confirmed in correlation tests (Spearman’s rho correlation coefficient= -0.02; 95%CI, -0.24-0.25;P = .99). However, a weak yet significant correlation was seen between Q and hyperemic MTT (Spearman’s rho= -0.28; 95%CI, -0.01-0.51;P = .04), and between R and IMR (Spearman’s rho= 0.28; 95%CI, 0.04-0.51;P = .03).

Complications

While the vasoreactivity test was being performed, 3 cases of transient bradycardia (5%) without clinical repercussions and 2 episodes of atrial fibrillation (4%) were reported, 1 of them self-limited while the other required sedation and electrical cardioversion. After the administration of regadenoson, most patients experienced some degree of discomfort, which was well-tolerated and reversed with the administration of 100 mg of intravenous theophylline. No other complications or adverse effects were reported.

DISCUSSION

This study confirms that a high percentage of patients with symptoms or signs of INOCA show microvascular dysfunction or vasospasm in invasive functional testing, and that it is feasible and safe to perform (figure 5).

Figure 5. Study design, endotype-based classification, and analysis using the ROC curve. AUC, area under the curve; CFR, coronary flow reserve; IMR, index of microvascular resistance; INOCA, ischemia with nonobstructive coronary artery disease; Q, absolute coronary blood flow; R, microvascular resistance.
* P < .05.

 

The percentage of patients with microcirculation or vasomotility alterations found in our study (77%) is consistent with former studies of patients with angina without obstructive coronary artery disease (64% to 89%8-11).

Vasoreactivity test

Some groups systematically use a dose of 200 µg of intracoronary acetylcholine to perform the vasoreactivity test; in our study, the high dose was established at 100 µg according to the COVADIS group, the CorMicA protocol, and the technical document of the Spanish Society of Cardiology Working Group on Cardiac Catheterization and Interventional Cardiology, which highlights its high sensitivity and specificity rates (90% and 99%, respectively).12 As a matter of fact, the high prevalence of positive results seen in our study in the acetylcholine test (60%) is similar to that reported in other series (57% to 71%13-15). In a recent study of 110 patients, Feenstra et al.11 revealed that 62% of the patients had a pathological acetylcholine test that confirmed the presence of epicardial vasospasm and microvascular spasm (36% and 26%, respectively).

In our study, the complications associated with the vasoreactivity test in our study are not very many: 2 cases of atrial fibrillation (4%), which is consistent with the incidence rate reported by the CorMIcA trial (5%).9

Prevalence of endotypes

The most common endotype in our patients was isolated microvascular dysfunction (40%), but not as much as in the CorMicA trial (52%). These differences could be explained by the discrepancy seen in the percentage of completely normal angiographies (22% in the CorMicA vs 63% in our study) due to the possible association between non-obstructive atheromatous disease and microvascular dysfunction.16,17

The prevalence of the remaining endotypes is similar to that reported in the CorMicA trial: isolated epicardial vasospasm (17% vs 17%), and mixed disorder (20% vs 21%). A recent meta-analysis that included 14 427 patients with INOCA also shows similar percentages.18

Indices of microvascular function obtained through bolus thermodilution

The analysis of the MTT obtained with this technique (figure 2), a parameter that correlates inversely with the direct measurement of coronary flow,19 reveals an interesting finding that is consistent with the data published by Nardone et al.20: patients with low CFR have 2 differentiated phenotypes based on the IMR. On the one hand, cases with reduced CFR and elevated resistances have normal baseline flow and low hyperemic flow, which would be indicative of an insufficient vasodilation response. However, in patients with normal resistances, a reduced CFR would be indicative of an abnormally elevated resting flow with hyperemic flow in the normal range. This phenomenon can also be observed in the analysis of patients with high Q (table 3) in whom a reduced CFR can be attributed to elevated baseline flow instead of an insufficient hyperemic response.

Therefore, this subgroup probably shows inefficient or dysregulated baseline myocardial flows. This characteristic, of indeterminate cause, could have important therapeutic implications like a lack of response to vasodilator drugs.

Indices of microvascular function obtained by continuous thermodilution

The continuous thermodilution technique has evolved to the point of quantifying Q and R with a microcatheter and specific software in a simple and precise fashion. The main advantages of this method are its independence from an operator, reproducibility, and induction of hyperemia with a physiological saline solution without the need for pharmacological agents.21-24 However, its main limitation is the lack of normal reference values.

In our study, the lack of a correlation between Q and CFR could be justified by the variations described of baseline myocardial flow. Estimating the CFR requires estimating the baseline coronary flow while Q is a measurement that is representative of hyperemic flow.

The weak concordance seen in this study between Q and hyperemic MTT and between R and IMR shows how difficult it is to establish valid cut-off values for patient comparison with these indices.

With an optimal cut-off value of R in our study of 435 WU (an 81% sensitivity, and a 57% specificity), a total of 66% of cases with IMR ≥ 25 were properly categorized with this index. This value is somewhat lower compared to the one shown by Rivero et al.,25 who analyzed 120 patients and found that an R > 500 WU properly categorized 80% of the cases with IMR ≥ 25.Konst et al.26 studied 84 patients with INOCA using both thermodilution techniques only to find no correlation between the Q-R combo and IMR.

The differences seen may be explained by the fact that the quantitative variability of Q and R values among individuals mostly depends on myocardial mass. However, in positron emission tomography studies, considerable ranges were seen even after adjusting for flow and resistance values for myocardial mass. Therefore, it has been speculated that the most plausible hypothesis is the natural variation of hyperemic myocardial perfusion among individuals.27

Therefore, indices like CFR estimated by continuous thermodilution and microvascular resistance reserve are currently in the pipeline. They correlate the absolute values of flow and resistance seen during hyperemia with those obtained at rest. Nonetheless,these new parameters will still need validation in futurestudies.28,29

Limitations

The data presented here should be interpreted while understanding that this is an observational, single-center study with a small sample size. Therefore, results may be biased by confounding factors associated with a study of this nature.

The left anterior anterior descending coronary artery was considered as the pre-specified target vessel. However, in the routine clinical practice, it may be appropriate to assess other arteries in the presence of negative tests and high clinical suspicion.1

The optimal sequence in invasive functional studies has not been established yet.1 In our case, we chose to perform the acetylcholine test first to minimize the instrumentation of the artery and avoid further guidewire-induced vasoreactivity. However, the spasm and symptoms seen during the provocation test, although transient, could interfere with subsequent measurements of microvascular function. The possibility of determining CFR by continuous thermodilution was established at the beginning of our study, and it was assumed that a comparison of the CFRs obtained with both techniques would have been more appropriate.

In most bolus thermodilution studies, intravenous adenosine is used to induce hyperemia. However, we chose regadenoson because it is easy to use, following our previous experience, and because evidence says it is equivalent to adenosine.30,31

Finally, we should not overlook that this is an invasive study so potential risks associated with the examination should be weighed in. To this date, however, conducting this study has not impacted prognosis.

CONCLUSIONS

The invasive study of coronary vasoreactivity and microcirculation is feasible and safe. These studies allow us to easily recognize different endotypes of patients with INOCA and help us optimize their treatment.

The analysis of CFR, IMR, and Q combined can unmask a subtype of microvascular dysfunction characterized by an abnormally high baseline coronary flow.

The new indices obtained by continuous thermodilution show low concordance with respect to the reference indices. Therefore, future studies will be required to determine the utility of this technique.

FUNDING

None whatsoever.

AUTHORS’ CONTRIBUTIONS

All the authors contributed substantially to the study idea, design, and data mining process. In addition, all approved the manuscript final version for publication.

CONFLICTS OF INTEREST

None reported.

REFERENCES