Percutaneous treatment of the left main coronary artery in older adults. Impact of frailty on mid-term results

INTRODUCTION

The use of the transradial approach for coronary interventions has become increasingly widespread in interventional cardiology due to its numerous advantages.1 As a result, current guidelines recommend it as the first-line approach.2

However, the benefits of this technique are tempered by the risk of radial artery occlusion (RAO), with reported rates ranging from 5% to 30%.3,4 The aim of this study was to assess the incidence and predictors of RAO following transradial catheterization using Doppler ultrasound for evaluation.

METHODS

Patient population

This longitudinal, single-center prospective study was conducted in the cardiology department of the Military Central Hospital in Algiers. After applying exclusion criteria (hemodynamic instability and ST-segment elevation myocardial infarction), we included 427 consecutive patients undergoing transradial coronary procedures between January 2019 and March 2020. The study adhered to the principles of the Declaration of Helsinki and the International Conference on Harmonization Good Clinical Practices and was approved by the local ethics committee. All patients provided written informed consent.

Radial artery cannulation and retrograde radial arteriography

After radial artery (RA) puncture, a radial hydrophilic sheath (Radiofocus II, TERUMO Medical, Japan, or Prelude, MERIT Medical, United States) was introduced. An antispastic cocktail was then administered into the RA through the sheath, consisting of a saline solution, a vasodilator (1 mL of nicardipine), and a bolus of unfractionated heparin, which was administered either intravenously or directly into the RA as part of the spasmolytic cocktail, depending on the operator’s preference. In patients on vitamin K antagonists, these medications were not discontinued prior to the procedure.

Retrograde radial arteriography was performed by injecting a mixture of 4 mL of contrast and 4 mL of isotonic saline through the sheath. Radiographic images were then obtained in an anteroposterior projection.

Transradial coronary procedure

The standard approach was conventional right radial access. For coronary angiography, 5-French (Fr) hydrophilic sheaths and catheters were usually used. If the patient required revascularization, an ad hoc percutaneous coronary intervention was performed, using 6-Fr guiding catheters after exchanging the sheath from 5-Fr to 6-Fr. The usual dose of heparin is 5000 IU (2500 IU for oral anticoagulation with a vitamin K antagonist).

Hemostasis procedure

At the end of the procedure, the sheath was removed, and hemostasis was achieved using a hemostatic compression device (TR BAND, TERUMO Medical, Japan). A reverse Barbeau test5 was systematically performed. The hemostasis device was removed by nurses in the hospitalization unit. No standardized protocol for the duration of hemostasis was followed.

Assessment of postprocedural radial artery patency

Radial Doppler assessments were conducted before and after each transradial procedure. To evaluate RAO, pulsed Doppler was performed bilaterally on the radial and ulnar arteries. Normal arterial flow was indicated by a biphasic or triphasic signal, reflecting good perfusion. In cases of RAO, 2 additional ultrasonographic examinations were performed at 1 and 3 months, following the same protocol. Artery patency was assessed by an independent operator.

Classifications and definitions

RAO was defined as the absence of anterograde flow in the RA on ultrasound (figure 1). The location of the radial occlusion was identified using color and pulsed Doppler. We delineated 3 anatomical territories: the distal third, extending from the radial styloid to approximately 7 to 10 cm proximally; the proximal third, from the elbow folds to approximately 7 to 10 cm distally; and the middle third, located between the previous 2 regions (middle part of the forearm).

Figure 1. Radial artery with occlusion in the distal third. Pulsed Doppler flow targets a stop flow indicating radial occlusion.

The type of hemostasis, whether occlusive or patent, was assessed: patent hemostasis was indicated by the presence of a plethysmographic signal in the RA during the reverse Barbeau test,5 which involves compression of the ulnar artery. The operator did not intervene during this process but simply recorded whether the artery remained patent or not.

The internal luminal diameter of the RA was defined as the distance between the leading edges of the intima-lumen interface on the superficial wall and the lumen-intima interface on the deep wall.6

The R/S ratio (radial/sheath) was calculated by dividing the luminal diameter of the RA by the external diameter of the sheath (Radiofocus II: 5-Fr = 2.29 mm, 6-Fr = 2.62 mm, 7-Fr = 2.97 mm; Prelude: 5-Fr = 2.52 mm, 6-Fr = 2.83 mm). This ratio was categorized qualitatively as < 1 or ≥ 1.

RA anatomical variations of clinical relevance were classified according to definitions provided in the literature.7,8 A high origin (high bifurcation) of the RA (figure 2) was defined with reference to the intercondylar line of the humerus. A radioulnar loop was characterized by the presence of a complete 360° loop of the RA, while radial tortuosity was identified by a curvature greater than 45°.

Figure 2. Anatomic variations of the radial artery. A: high origin of the radial artery. The radial and ulnar arteries separate at the level of the middle third of the humerus (arrow). B: radioulnar loop was defined as a complete 360° loop of the radial artery distal to the bifurcation of the brachial artery (arrow).

A blood pressure profile was obtained on the same side as the radial access. Forearm hematomas were classified according to the “EASY” study9: type I: < 5 cm in diameter; type II: < 10 cm; type III: > 10 cm but not extending to the elbow; type IV: extending beyond the elbow; type V: resulting in an ischemic lesion.

Statistical analysis

The statistical analysis was performed using IBM SPSS Software version 25. Parameters of interest are reported with their 95% confidence intervals (95%CI). For all tests, a significance threshold of 5% was retained. All tests were performed bilaterally. The following tests were used to compare groups: the chi-square test was used to compare 2 qualitative variables; the Student t-test or analysis of variance was used to compare a quantitative variable with a qualitative variable, with the Fisher test being applied when variances were unequal; and logistic regression was used to identify predictors of RAO.

RESULTS

Clinical and procedural characteristics of the study population

During the study period, 441 patients were screened. Of these, transradial access failed in 14 patients, who were excluded from the study, resulting in an eligible sample of 427 patients (mean age 61.9 ± 11.1 years, 67.4% male). Among the patients, 260 had hypertension (60.9%), and nearly half had diabetes (48.9%).

Table 1 summarizes the procedural data. The sheaths used were mainly 6-Fr (83.6%), and heparin was injected intra-arterially in 63.5% of patients. The mean heparin dose was 5669 ± 1394 IU, with a higher dose given when percutaneous coronary intervention was performed (4940 ± 339 IU vs 7491 ± 1368 IU; P < .001).

Table 1. Procedural data

Incidence and characteristics of radial artery occlusion

RAO occurred in 48 patients (11.24%). Of these, 89.6% were asymptomatic, and the radial pulse remained palpable in 14 patients (29.2%). At 1 month, 2 patients were lost to follow-up. Among the remaining 46 patients, spontaneous recanalization occurred in 13 patients (28.3%). At the 3-month follow-up, the recanalization rate increased to 32.6% (15 cases).

The site of RAO was the distal third in 7 patients (14.6%), the middle third in 21 patients (43.8%), and the proximal third in 20 patients (41.7%).

Predictors of radial artery occlusion

Patients with RAO were significantly younger (table 2). The mean periprocedural systolic blood pressure in the RAO group was significantly lower (138.04 mmHg ± 21.92, vs 145.84 mmHg ± 21.10; P = .017). Type A Barbeau test was associated with a higher risk of RAO compared with types B and C, and patients with occlusion had a smaller RA diameter (2.34 mm ± 0.40 vs 2.61 mm ± 0.37; P < .001) (figure 3).

Table 2. Comparison of patients with and without RAO

Figure 3. Radial artery diameter as a predictor of occlusion. A: the radial diameter is significantly smaller if there is RAO. B: less than 2.5 mm, the risk of occlusion becomes greater.

RAO procedural factors are listed in table 2. An R/S ratio < 1 was found in 35 patients in the RAO group vs 153 patients in the non-RAO group (72.9% vs 40.3%, P < .001). The mean heparin dose was significantly lower in patients with RAO (5007 ± 1352IU vs 5754 ± 1378 IU; P < .001), and the dose adjusted to weight was also significantly lower in the RAO group (62.31 ± 17.82 IU/kg vs 75.73 ± 22.57 IU/kg; P < .001). In addition, the RAO rate decreased significantly when the heparin dose exceeded 70 IU/kg.

Forty-two patients in the RAO group had occlusive hemostasis vs 241 in the non-RAO group (87.5% vs 63.5%; P = .001). Surprisingly, two-thirds of our patients (283 [66.3%]) had occlusive hemostasis. The mean duration of hemostasis was longer if there was RAO (5.15 h ± 1.41 vs 4.29 h ± 1.22; P < .001).

On multivariate logistic regression analysis (figure 4), the following factors were independent predictors of RAO: young age (odds ratio [OR], 0.642; 95%CI, 0.480-0.858; P = .031), low periprocedural systolic blood pressure (OR, 0.598; 95%CI, 0.415-0.862; P = .007), type A Barbeau test (OR, 0.441; 95%CI, 0.198-0.981; P = .045), small RA diameter (OR, 0.371; 95%CI, 0.323-0.618; P = .031), insufficient anticoagulation (OR, 0.287; 95%CI, 0.163-0.505; P < .001), occlusive hemostasis (OR, 0.128; 95%CI, 0.047-0.353; P < .001), and a long hemostasis duration (OR, 1.786; 95%CI, 1.428-2.039; P < .001).

Figure 4. Independent factors predictive of radial artery occlusion. Multiple logistic regression analysis revealed that the independent factors predictive of radial occlusion were young age, low periprocedural systolic blood pressure, type A Barbeau test, small radial artery diameter, insufficient anticoagulation, occlusive hemostasis, and long hemostasis duration. 95%CI, 95% confidence interval; PSBP, periprocedural systolic blood pressure; RAO, radial artery occlusion.

Anatomic variations of the radial artery

The mean radial diameter was 2.58 mm ± 0.39, and the diameter was larger in men (2.69 mm ± 0.37 vs 2.36 mm ± 0.31; P < .001) and smaller in patients with diabetes (2.53 mm ± 0.38 vs 2.64 mm ± 0.38; P = .003). The mean radial diameter was significantly larger than the mean ulnar diameter (2.58 mm ± 0.39 vs 2.22 mm ± 0.43; P < .001).

Radial anatomical variations affected 63 patients (14.8%). The most common variation was a high origin of the RA, observed in 63.5% of cases (40 patients), followed by radial tortuosity in 28.6% (18 patients), radioulnar loop in 7.9% (5 patients). Anatomical variations were more frequent in women (23% vs 10.8%; P = .001) and in older patients, with a mean age of 66.3 years ± 10.2 vs 61.2 years ± 11.2 in those without variations (P = .001).

Periprocedural complications

Radial spasm occurred in 72 patients (16.9%). This complication was more frequent in women (29% vs 10.1%; P < .001), patients with diabetes (22.5% vs 11.5%; P = .002), and when 6-Fr catheters were used (14% vs 24%; P = .035). Forearm hematoma occurred in 25 patients (5.85%). According to the EASY classification,9 most hematomas were type I (17 patients, 68%), followed by type II (6 patients, 24%), with type III occurring in only 2 patients (8%).

DISCUSSION

The rate of RAO remains relatively high in some institutions.10,11 In the PROPHET study, the acute incidence of RAO (12%) was almost halved in 28 days (7%).3 Recanalization occurs as a result of activation of primary fibrinolysis.12 In the present study, the rate of radial recanalization at 3 months was 32.6%. The only predictor of recanalization was radial diameter: the larger the diameter, the higher the rate of spontaneous recanalization.

Zankl et al.13 found that RAO was located in the distal third of the forearm in 49% of patients, in the distal and middle third in 13.7%, and in the entire forearm (proximal third) in 37.3%. Dissections of the media also occur in the proximal RA, likely due to catheter progression or manipulation without protection of the sheath.14 In our opinion, this would explain the location of RAO in the proximal part of the artery.

Among our patients with RAO, 29.2% had a radial pulse. According to Uhlemann et al.,4 in 19.5% of patients with RAO on Doppler, the RA pulse was still palpable. This was likely due to retrograde filling of the RA by collaterals. Therefore, the diagnosis of RAO should be confirmed using a more objective method, such as Doppler ultrasound.

Young age is a predictor of RAO, possibly due to higher sympathetic reactivity in younger individuals, which increases their risk of spasm. However, this characteristic does not influence the rate of recanalization, likely because prolonged radial spasm leads to the formation of a permanent intra-arterial thrombus.

Low mean systolic blood pressure was also a predictor of radial occlusion. We speculate that hypertension and arterial stiffness may prevent complete interruption of flow during compression, thereby helping to maintain radial patency.15

There was a higher incidence of occlusion with type A Barbeau test. We believe that in cases with well-developed ulnar circulation, the ulnar artery generates a competitive retrograde flow that opposes the radial flow, promoting occlusion and hindering recanalization.

The likelihood of developing RAO is related to the size of the sheath,16 or more precisely, the R/S ratio.17 A prospective registry showed that 5-Fr sheaths reduced the rate of RAO by up to 55% compared with 6-Fr.4

A study by Pancholy et al.,18 demonstrated that intravenous heparin is as effective as intra-arterial heparin in reducing the incidence of RAO, suggesting that the systemic effect of heparin is more important than its local effect. A recently published meta-analysis identified higher heparin doses as the most significant measure for decreasing RAO.12 This results is in line with our finding that a dose of less than 70 IU/kg seems to promote the occurrence of RAO. The high prevalence of RAO and the benefit of higher doses of unfractionated heparin (≥ 50 IU/kg) in this setting were also highlighted by a meta-analysis of 112 studies.19 In a randomized superiority trial comparing high-dose (100 IU/kg) and standard-dose (50 IU/kg) heparin, the RAO rate was significantly lower in the high-dose group.20 Recent evidence suggests that a small dose of rivaroxaban, given orally after a transradial procedure, may decrease the occurrence of RAO at 1 month.21,22

Using the reverse Barbeau test, Sanmartin et al.23 found that 60% of patients had an absence of radial flow during compression. These observations led to the concept of nonocclusive hemostasis (patent hemostasis). In the PROPHET study,3 RAO was significantly less frequent in the group that underwent nonocclusive hemostasis than in the control group.

The duration of hemostatic compression has been studied in large, randomized trials.24-26 The authors concluded that compression duration was a strong predictor of RAO.

In a meta-analysis by Rashid et al.,27 the incidence of RAO after diagnostic coronary angiography was notably higher compared with percutaneous coronary intervention, possibly due to the use of higher anticoagulation doses during interventions.12 However, opposite findings have been reported by other studies.

In our sample, the mean radial diameter was 2.58 mm ± 0.39 and was significantly larger in men. Velasco et al.28 reported a mean arterial diameter of 2.22 ± 0.35 mm, while a Polish study found a mean diameter of 2.17 ± 0.53 mm for the right RA and 2.25 ± 0.43 mm for the left RA.29 The ulnar artery is also used in interventional cardiology,30 although there is no consensus on its size compared with the RA.

Autopsy studies of arterial anatomic variations of the upper extremity have reported frequencies between 4% and 18.5%.8 In the literature, the most frequent anatomic variation of the RA is high bifurcation. Yoo et al.31 reported a 2.4% incidence of high radial origin in 1191 Korean patients. Tortuosity of the RA frequently affects patients with high radial origin, possibly due to the elongated course of the RA predisposing it to tortuosity, which is considered one of the most common causes of procedural failure, along with radial spasm.32

Radioulnar loop is the most common cause of procedural failure with experienced operators.33 Angiographic evaluation of the radioulnar anastomosis is mandatory in such cases, as there is often a negotiable anastomosis between the radial and ulnar arteries.

In our study, radial spasm was the leading cause of procedural failure, occurring in 50% of the 14 patients who experienced such failures. Ruiz-Salmerón et al.34 found that RA anatomic variations were strongly associated with radial spasm in a multivariate analysis. The relationship between radial spasm and anatomic variations is mainly explained by the strong correlation with high radial origin and the radioulnar loop.

Study limitations

Since this study is a prospective registry and not a randomized trial, selection bias cannot be excluded. Our study represents a single-center experience with a limited number of patients, despite being one of the largest prospective registries of vascular ultrasound in radial catheterization to date. Among the other limitations of the study, we note the lack of standardized protocols for both heparin use and compression.

CONCLUSIONS

With the increasing number of transradial procedures and the greater age of patients undergoing these interventions, leading to more complex procedures, it is essential to maintain the patency of the RA for future access. Although predictors of RAO after cardiac catheterization have been identified, implementing preventive measures in practice remains a challenge. The main modifiable predictors associated with the risk of RAO are insufficient heparinization and occlusive hemostasis. Therefore, preventive strategies should primarily focus on addressing these 2 factors.

FUNDING

None.

ETHICAL CONSIDERATIONS

The study was conducted in accordance with the provisions of the Declaration of Helsinki and with the International Conference on Harmonization Good Clinical Practices and was approved by the local ethics committee. All patients included in the study provided written informed consent, which is archived and available. Our study population included both sexes. Gender had no influence on the occurrence of radial occlusion.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

No artificial intelligence software was used in the preparation of this study.

AUTHORS’ CONTRIBUTIONS

All authors meet the criteria for authorship as defined by the International Committee of Medical Journal Editors. M.S. Lounes, A. Meftah, C. Belhadi, K. Allal, H. Boulaam, A. Sayah, I. Hafidi, and E. Tebache contributed to the acquisition and analysis of data for this article. M.S. Lounes, A. Bedjaoui, A. Allali, and S. Benkhedda were responsible for the study design and the writing of the article. M.S. Lounes, A. Allali, and S. Benkhedda contributed to writing and critical revision of the content. All authors have read and approved the final version of the article and agree to be accountable for all aspects of the work, including the accuracy and integrity of all its parts.

CONFLICTS OF INTEREST

None.

REFERENCES

INTRODUCTION

The left main coronary artery (LM) supplies 84% of the blood flow to the left ventricle in patients with right dominance,1 making LM disease the coronary lesion with the worst prognosis. The prevalence of this disease is not negligible, as it is found in 4.8% of coronary angiograms,2 highlighting the prognostic importance of these lesions. Conservative treatment is a rarely a feasible option due to the high rate of cardiac adverse events during short-term follow-up, with a mortality rate exceeding 50%.3

Coronary artery bypass grafting (CABG) has traditionally been the most widely accepted revascularization strategy.4 In recent years, there have been significant pharmacological and technological improvements in percutaneous revascularization techniques, such as drug-eluting stents and intracoronary diagnostic techniques.5 These improvements, together with comparative studies, have prompted discussion on the various alternatives.6 Presently, the choice of revascularization strategy should be based on the complexity of the coronary anatomy and surgical risk.7

However, evidence is limited in older adults who are scarcely represented in classic studies. Furthermore, in these patients, frailty is a frequent and unstudied characteristic that can influence their prognosis. In this special population, CABG is usually ruled out due to high-surgical risk. On the other hand, percutaneous coronary intervention (PCI) could be a potential therapeutic option, although with little evidence to date.8 Consequently, we postulated that PCI of the LM might be feasible and safe in older patients, with a low incidence of associated complications and an acceptable rate of major adverse cardiac events (MACE) during follow-up.

METHODS

Study design

We conducted a retrospective, single-center study of older patients diagnosed with LM disease who underwent PCI. The study aimed to evaluate mid-term clinical outcomes and examine the prognostic significance of frailty in these patients. The study protocol was approved by the local clinical research ethics committee according to institutional and good clinical practice guidelines. Recruitment took place from January 2017 to December 2021 at Hospital Universitario Reina Sofía (Cordoba, Spain). Patients were eligible if they were aged ≥ 75 years at the time of LM disease diagnosis, and PCI was chosen as the treatment after deliberation by heart team discussion, or due to instability requiring emergent revascularization. Exclusion criteria consisted of end-stage chronic diseases, patients under palliative care, contraindications to dual antiplatelet therapy, and incomplete follow-up data. Included patients were grouped according to frailty status, determined by the FRAIL scale, with patients scoring 3 or more points considered frail.9 Definitions are shown in the supplementary data.

Outcomes

The main objective of the study was to describe mid-term clinical outcomes in older patients undergoing LM-PCI. We also aimed to compare clinical events according to the presence of frailty. The primary endpoint was a composite of MACE, defined as a composite of cardiovascular death (including death of uncertain cause), nonfatal myocardial infarction, the need for new revascularization, and stroke. Secondary outcomes were the individual components of MACE and all-cause mortality.

Angiographic analysis

Quantitative analysis of the coronary arteries was performed using the validated CAAS system (Pie Medica Imaging, the Netherlands). The basal anatomy of the LM bifurcation with the anterior descending artery and the circumflex artery was classified according to the Medina classification.10 The measurements analyzed included the reference diameter of the LM and its percentage of stenosis. The complexity of the coronary anatomy was studied using the SYNTAX scale.6

Statistical analysis

Categorical data are presented as counts (percentages), while continuous data are expressed as mean ± standard deviation or median [interquartile range]. Between-group comparisons were performed using the chi-square test or the Fisher exact test for categorical variables and the Student t-test or the Mann-Whitney U test for continuous variables. Kaplan-Meier curves and Cox regression models were used to analyze clinical events according to frailty. Inverse probability of treatment weighting (IPTW) was used to account for clinical differences between the 2 groups.11 Propensity scores were calculated using a logistic regression model that included the following covariates: age, sex, left ventricular ejection fraction, atrial fibrillation, chronic kidney disease, anemia, and chronic obstructive pulmonary disease. Standardized mean differences before and after weighting were used to evaluate the balance of the groups regarding the covariates. A difference of < 10% was considered to indicate a satisfactory balance. The distributions of the propensity scores before and after weighting were plotted to assess the degree of overlap between the 2 groups. Confidence intervals for the IPTW coefficients were obtained using robust sandwich-type variance estimators (figure 1 of the supplementary data).12 All tests were 2-tailed and significance was set at P < .05. Statistical analyses were performed using SPSS software (V 24; IBM Corp., United States) and R software (V4.0.3; R Foundation for Statistical Computing, Austria).

Figure 1. Main events to follow-up. CV, cardiovascular; MACE, mayor adverse cardiovascular events; MI, myocardial infarction; NS, nonsignificant; PCI, percutaneous coronary intervention. * P < .005.

RESULTS

During the study period, our hospital treated 437 patients with significant LM lesions percutaneously. Of them, a total of 140 patients met the inclusion criteria and were included in the analysis (figure 2 of the supplementary data).

Figure 2. Kaplan-Meier Curves of the primary outcome and mortality. CV, cardiovascular; MACE, major adverse cardiovascular events.

Baseline characteristics

The baseline clinical characteristics, clinical presentation and antithrombotic treatment administered are detailed in table 1. The median age of the patients was 80 [78-84] years and 36% (51 patients) were women. Most of the patients had a history of hypertension (84%, 118 patients) and 58% (81 patients) were diabetic. More than a third of the patient cohort had a previous personal history of ischemic heart disease (37%, 52 patients) and 33% (46 patients) had chronic kidney disease. Among noncardiovascular comorbidities, active cancer was present in 11 patients (8%) and prior blood transfusions had been required in 16 patients (11%). The mean EuroSCORE II was 3.07 [1.96-5.7] to assess surgical risk. Forty-eight patients (34%) had left ventricular systolic dysfunction at the time of revascularization.

Table 1. Patients’ baseline characteristics

The most common clinical presentation was acute coronary syndrome (85 patients, 61% of cases). Among these, onset consisted of ST-segment elevation myocardial infarction (STEMI) in 9 patients (6%), non-ST-segment elevation myocardial infarction in 61 patients (44%), and unstable angina in 15 patients (10%). The remaining patients (55, 39%) presented with chronic coronary syndrome.

A total of 104 patients (74%) were discharged with dual antiplatelet therapy. The main combination was aspirin and clopidogrel (61 patients, 43%). In 36 patients (26%), initial triple therapy (anticoagulation and dual antiplatelet therapy) was chosen due to concurrent conditions requiring chronic oral anticoagulation.

Based on the FRAIL scale, almost half of the patients (68 patients, 49%) met clinical criteria for frailty at the time of revascularization. The baseline characteristics of frail and nonfrail patients are shown in table 1. No statistically significant differences were found in terms of age, main cardiovascular risk factors or noncardiovascular comorbidities between the 2 groups. However, compared with nonfrail patients, those with frailty were more likely to be female (49% vs 25%; P = .004), to have atrial fibrillation (22% vs 10%; P = .041), a higher EuroSCORE level (3.80 vs 2.76; P = .010), and anemia (28% vs 14%; P = .040), and consequently a lower hematocrit and hemoglobin value (36.6% vs 39.6%; P = .031 and 12.16 mg/dL vs 13.02 mg/dL; P = .017, respectively).

Angiographic and procedural characteristics

Angiographic and procedural data are shown in table 2. The arterial access of choice was radial access (81% of procedures, 113 patients). A median SYNTAX score of 21 [15-29.5] was observed in 96 patients (68%) with multivessel disease, and 62 patients (44%) had a SYNTAX score > 22. The most common angiographic involvement of the LM was the distal segment (61%, 86 patients), while the most common plaque distribution according to the Medina classification was “1,1,1” (35 patients, 41% of LM bifurcation lesions). The strategy of choice for the treatment of the bifurcation was the provisional stent strategy (85% of LM bifurcation lesions, 73 patients), while the upfront 2-stent strategy was used in only 13 patients (15% of the LM bifurcation lesions). The mean diameter of the LM was 4.1 [± 3.5-4.5] mm with a mean angiographic stenosis of 62% (± 7). In 59 patients (42%), the procedure was guided using intravascular imaging techniques (58 patients using intracoronary ultrasound and 1 patient using coherence tomography). Coronary physiology was used in 5 patients (4%) to guide the need for revascularization or to check the result after percutaneous treatment. In 7 (5%) patients, mechanical support was required, either due to cardiogenic shock, or as a preventive measure in high-risk angioplasty (5 patients with an intra-aortic balloon pump and 2 with an Impella CP device [Abiomed, United States]). Intraprocedural complications occurred in 8 patients (6%), including a major complication in 4 patients (3 intraprocedural deaths and 1 cardiogenic shock), and a minor complication in 4 patients (1 coronary dissection with Thrombolysis in Myocardial Infarction (TIMI) grade 3 distal flow, 1 pseudoaneurysm, and 2 bleeding events from the femoral access resolved by stent implantation). The LM diameter was larger in patients with frailty than in those without (4 mm [4-4.5] vs 3.5 mm [3.5-4.5]; P = .023), a paradoxical finding since the percentage of women was higher in the group with frailty percentage of women. However, this information did not seem to be clinically relevant. No other clinically relevant differences were found between the 2 groups (table 2).

Table 2. Patients’ angiographic and procedural characteristics

Clinical results at follow-up

After a median follow-up of 19 months [5-35], a total of 40 (29%) MACE were recorded: 3 (2%) patients had a nonfatal myocardial infarction, 7 (5%) patients required repeat revascularization (3 for restenosis of the LM, and 4 in a different vessel), and 30 patients (21%) died of cardiac and/or uncertain causes. No strokes were reported during follow-up. Sixteen patients (11%) died of noncardiac causes during follow-up.

Clinical outcomes are presented in figure 1 and figure 2. No independent predictor of MACE was identified. The independent predictors of all-cause mortality were left ventricular ejection fraction (hazard ratio [HR], 0.90 [0.96-0.99]; P = .014), chronic kidney disease (HR, 2.26 [1.16-4.42]; P = .017), and particularly the presence of frailty (HR, 2.42 [1.17-5.02]; P = .018) (table 1 of the supplementary data). The primary endpoint of MACE occurred in 24 (35%) patients in the frail group and in 16 (22%) patients in the nonfrail group (HR, 1.61 [0.79-3.28]; P = .193). Frail patients had an increased risk of cardiovascular mortality: 21 (31%) vs 9 (13%); HR, 2.64 (1.21-5.77); P = .015. All-cause mortality was also more frequent in the frail group: 33 (49%) vs 13 (18%); HR, 2.94 (1.55-5.59); P = .001). The events during follow-up are presented in table 2 of the supplementary data. After IPTW adjustment, only the difference in all-cause mortality remained significant (HR, 1.95 [1.02-3.75]; P = .046). Survival analysis of the weighted population is shown in figure 3.

Figure 3. Kaplan-Meier Curves of the secondary outcomes. CV, cardiovascular; IPTW, inverse probability of treatment weighting; MACE, major adverse cardiovascular events.

DISCUSSION

The present study describes the feasibility of LM-PCI in a cohort of older patients. The main results were as follows: a) the rate of MACE at mid-term follow-up was 29%, mainly driven by cardiovascular and/or uncertain cause death; b) a high percentage of frailty was found in our population (49%); c) frail patients had a 2-fold increased risk of all-cause mortality during follow-up (HR, 1.95 [1.02-3.75]; P = .046) (figure 4).

Figure 4. Central illustration. Results of percutaneous treatment of LM in elderly patients and impact of frailty. CV, cardiovascular; LM, left main coronary artery; MACE, major adverse cardiovascular events; MI, myocardial infarction; NS, non-significant; PCI, percutaneous coronary intervention.

The treatment of LM disease has traditionally been surgical, given the complexity involved and significant prognostic impact.13 However, the marked advances in interventional cardiology in recent decades have modified the approach.14,15 Contrasting evidence from clinical trials and meta-analyses shows that percutaneous treatment has similar results to surgical approaches in terms of mortality, acute myocardial infarction, and stroke at 5 years of follow-up.16 This shift has is reflected in the evolving recommendations in clinical practice guidelines, and the current European revascularization guidelines assign a grade of recommendation IA to both surgical and percutaneous strategies for the treatment of LM disease when the anatomy is not complex (SYNTAX < 22), and a class IIa recommendation for cases of intermediate complexity (SYNTAX 23-32).7

Nevertheless, the population analyzed in the study has specific clinical characteristics, and is not usually represented in large clinical trials (older patients and those with frailty and a high burden of associated comorbidities). These variables are not systematically included in surgical risk scores but are generally taken into account in routine clinical practice and often influence heart team decisions on the treatment strategy.17 Therefore, because this particular patient cohort is often excluded from research, there are no conclusive data on the benefit of percutaneous revascularization.

Our results are in line with those of previous registries in terms of MACE and all-cause mortality, as well as the association between age and a marked incidence of mortality due to noncardiac causes during follow-up. However, unlike earlier studies, we observed no differences in cardiovascular mortality, despite these patients having a more complex coronary anatomy than younger patients.18 In this regard, our study cohort had a median SYNTAX score of 21, and 44% of the patients had a score above 22. Like previous studies, this SYNTAX index score was not associated with a higher probability of cardiac events during follow-up in this special population.

In the present study, rates of acute myocardial infarction and new revascularization of the target lesion were lower than in other cohorts. Although it is difficult to make direct comparisons, we postulate that the use of new-generation drug-eluting stents and a higher proportion of revascularization guided by intracoronary diagnostic techniques may have influenced this finding. However, the use of intracoronary imaging techniques in our study was relatively low (42%) considering their benefit in patients with complex coronary lesions.19

In recent years, there has been growing interest in understanding the impact of comorbidities and frailty in older patients with cardiovascular disease.20,21 Several studies have compared invasive strategies with conservative approaches in older patients, demonstrating benefits for revascularization.22,23 However, the MOSCA-FRAIL trial compared both strategies in frail patients and observed that an invasive strategy did not confer additional benefit compared with conservative management of these patients, despite a fairly low percentage of LM disease.24 In our study, we observed a 2-fold increase in the risk of all-cause mortality in patients with frailty, suggesting the need to add systematic evaluation of frailty in older patients undergoing LM-PCI. Such assessment can aid in selecting the optimal therapeutic strategy, taking into account the likelihood of mortality during follow-up, irrespective of the application of an invasive strategy in coronary disease. These results, moreover, are consistent with other cardiovascular diseases with significant prevalence and mortality, such as heart failure.25

Study limitations

The present study has several limitations. First, it has the limitations inherent to its observational and retrospective design. Although the sample size is relatively small, it represents the largest study specifically focused on LM-PCI in older patients and analyses associated comorbidities and their impact on cardiovascular adverse events. Second, the absence of a control group receiving conservative treatment hinders the ability to draw more robust conclusions on the safety and efficacy of LM-PCI in these patients. In addition, the selection of cutoff points (age ≥ 75 years) to define this cohort of older patients was arbitrarily based on the exclusion criteria of the main clinical trials previously published. A high percentage of patients with frailty may not have undergone revascularization and would therefore have been excluded from the study. Regarding the prognostic significance of frailty, although we used IPTW to reduce confounding bias, we cannot rule out the possibility of residual confounding due to unmeasured covariables. Furthermore, there are no data on bleeding events during follow-up, which is an important concern given the impact of antiplatelet therapy in these patients. Finally, the percentage of intracoronary imaging use was lower than expected.

CONCLUSIONS

In real-life patients with advanced age and multiple associated comorbidities, percutaneous treatment of LM could be considered a feasible option, with an acceptable incidence of adverse cardiovascular events during follow-up and a low incidence of complications associated with the procedure. Frailty was an independent predictor of all-cause mortality during follow-up. When weighing the risks of LM-PCI in older patients, frailty should be taken into account in the therapeutic decision-making process.

FUNDING

None.

ETHICAL CONSIDERATIONS

The study protocol was approved by the Local Clinical Research Ethics Committee according to institutional and Good Clinical Practice guidelines. All patients signed the informed consent for publication. The authors confirm that sex and gender variables have been considered in accordance with the SAGER guidelines.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

No artificial intelligence was used in the preparation of the study.

AUTHORS’ CONTRIBUTIONS

I. Gallo, M. Alvarado and J. Perea contributed to data collection. R. González-Manzanares performed the statistical analysis. J. Suárez de Lezo and M. Romero contributed to the interpretation of the results. I. Gallo and F. Hidalgo wrote the manuscript. S. Ojeda and M. Pan reviewed the manuscript.

CONFLICTS OF INTEREST

S. Ojeda is associate editor of REC: Interventional Cardiology. The journal’s editorial procedure to ensure impartial processing of the manuscript has been followed. S. Ojeda has received consulting fees from Medtronic and Edwards and speaker fees from Philips, World Medical and Boston Scientific and is holder of a research grant (PI21/00949) from the Spanish Ministry of Science and Innovation (Instituto de Salud Carlos III). M. Pan has received speaker fees from Abbott, Boston Scientific, World Medical and Philips and holds a research grant (PI21/00949) from the Spanish Ministry of Science and Innovation (Instituto de Salud Carlos III). The remaining authors declare no conflicts of interest.

SUPPLEMENTARY DATA

REFERENCES