Article Text
PDF
Abstract
Objective To investigate whether treatment of lesions of greater complexity is now undertaken and to assess the rates of procedural success per class of lesion complexity.
Design Observational study.
Setting Despite impressive progress in treatment strategies and equipment, the success rate of percutaneous coronary intervention for chronic total occlusion (CTO) has remained relatively stable.
Participants 483 patients consecutively treated with CTO from 2003 to 2012.
Main outcome measures The Multicenter CTO Registry of Japan (J-CTO) score was used to classify lesion complexity. The study population was subdivided into an early (period 1, n=288) and a late (period 2, n=195) period according to the routine implementation of novel techniques and advanced equipment.
Results Period 2 was marked by more ‘difficult’ and ‘very difficult’ lesions (J-CTO grades 2 and 3) being attempted, with procedural success increasing from 68.4% to 88.1% (p<0.001) and from 42.0% to 78.9% (p<0.001), respectively. ‘Easy’ and ‘intermediate’ lesions (J-CTO grades 0 and 1) were less common, but with similarly high success rates (89.1% vs 96.6% (p=0.45) for easy, and 86.3% vs 86.1% (p=0.99) for intermediate). Period 2 was characterised by a trend for more successful procedures overall (by 6.1%, p=0.09). Procedural complications were similarly low in both periods. J-CTO score and technical era were identified as independent correlates of success in the total population by logistic regression analysis.
Conclusions Advanced CTO techniques and equipment have resulted in an increase in the successful treatment of highly complex lesions. Total success rate did not substantially improve, as it was counterbalanced by the increased rate at which complex lesions were attempted.
Statistics from Altmetric.com
Introduction
Few fields in interventional cardiology have seen more rapid progress in recent years than percutaneous recanalisation of chronic total occlusions (CTOs). Every aspect of the procedure, from guiding catheter support1 ,2 to the development of dedicated microcatheters and guidewires, and the implementation of new techniques (retrograde approach, knuckle wire, intravascular ultrasound guidance (IVUS)),3–5 has been improved to minimise some of the limitations that cause procedural failure. Very few studies have examined the effect on success rates consequent on this CTO revolution. Furthermore, as observed in previous longitudinal studies,6 the ability to demonstrate improved results with new techniques can be hindered by a trend to accept more complex lesions, sometimes resulting in worsening rather than improvement of the success rate over time. The mandatory publication of centres’ and operators’ results (as part of the freedom-to-access policy implemented for surgery and angioplasty in many countries including the USA and UK) has created more awareness of this problem,7 and adjustment methods have since been developed to correct outcome for patient and lesion selection. Therefore, the aim of this study was to analyse and compare CTO success rates with patient and lesion characteristics in a prospectively collected single-centre database of consecutive occlusions in the years 2003–2012.
Methods
Study population
This was an observational study conducted in a population of 483 consecutive patients with a native coronary artery CTO subjected to an attempt at percutaneous recanalisation between April 2003 and May 2012. Patients were selected on the basis of the presence of angina symptoms, viability of the myocardium subtended by the CTO artery, and inducible ischaemia in the CTO artery territory, as demonstrated by cardiac MRI, nuclear perfusion imaging, or exercise echocardiography. The exclusion criteria included an occlusion duration of <3 months, CTO of a coronary artery graft, and absent/low ischaemic burden in the CTO artery territory.
CTO was defined as a 100% coronary artery occlusion, with thrombolysis in myocardial infarction (TIMI) flow grade equal to 0 and duration >3 months.8 The occlusion duration was either angiographically proven or clinically estimated according to the start of symptoms or the timing of acute coronary events in the territory subtended by the CTO artery. Procedural success was defined as a residual stenosis <50% in the CTO lesion, with TIMI flow ≥ grade 2. The Multicenter Chronic Total Occlusion Registry of Japan (J-CTO) scoring system was used to assess lesion complexity, with classification of the lesions into easy, intermediate, difficult or very difficult9 (J-CTO score 0–3) according to presence of one or more of the following features: blunt stump, length >20 mm, severe calcification, >45° tortuosity, previous failed attempt.
We used the date of routine introduction of the Corsair microcatheter (Asahi Intecc, Nagoya, Japan) into our clinical practice to subdivide our study population into an early (period 1, 2003 to October 2008) and a late (period 2, November 2008 to May 2012) period. Table 1 details the main technical differences between the two periods. Along with the introduction of the Corsair microcatheter, period 2 was marked by the wide implementation of improved materials (eg, Fielder XT and Ultimate guidewires, 360 cm RG3 guidewires), dedicated devices (eg, CrossBoss catheter, Stingray catheter (BridgePoint Medical, Minneapolis, Minnesota, USA)) and technique (reverse controlled antegrade and retrograde subintimal tracking (CART), wire knuckling to facilitate wire progression).4
Procedural differences between period 1 (2003–October 2008) and period 2 (October 2008–2012)
Informed consent was obtained from all patients before their procedure in accordance with the current standards of practice.
CTO technique
All patients were treated with aspirin, clopidogrel and/or prasugrel before their procedure. When possible, arterial access was established via right and/or left femoral arteries. The size of the guiding catheters used for the occluded artery was preferentially 7 or 8 Fr. Dual injection was used routinely if contralateral collaterals were present, and a 7 Fr catheter was used also for the contralateral artery if a retrograde approach was considered. Long femoral sheaths were liberally used in the case of tortuous femoral and iliac arteries, with 45 cm sheaths routinely used in the last 2 years to allow better guiding catheter support and unhindered fine manipulation of the guidewires. An anchoring wire or balloon kept inflated in a small proximal vessel1 was used to provide extra support to the guiding catheter when required. An antegrade or retrograde approach was selected on the basis of the coronary artery anatomy, lesion characteristics and careful review of the dynamics of previously failed attempts. Single and parallel wire techniques and IVUS-guided identification of the entry point in stumpless occlusions were implemented in the antegrade approach. Controlled dissection techniques and distal re-entry facilitated with IVUS guidance5 or a Stingray4 device were used in the second period. When the retrograde approach was more extensively used after the routine introduction of the Corsair catheter,10 the CART technique11 was abandoned and retrograde crossing was achieved with a reverse CART technique, and subsequent wire insertion into the guiding catheter with wire externalisation. When appropriate (long lesion), a retrograde knuckled wire was used to rapidly reach the proximal end of the occlusion, avoiding perforation with stiff wires. The occluded segment was always stented with drug-eluting stents,12 ,13 and a balloon was used after dilatation to optimise stent expansion and apposition. During the procedure, the administration of unfractionated heparin maintained an activated clotting time of between 250 and 300 s, as monitored by blood sampling at regular intervals. Myocardial necrosis biomarkers were measured before and after the procedure, and percutaneous coronary intervention (PCI)-related myocardial infarction was defined as an increase in creatine kinase (with a positive MB fraction) greater than three times the upper normal limit.14 Coronary artery perforations were divided into major and minor, according to the requirement of pericardial drainage, interventional treatment with a covered stent or embolic device, or surgical treatment. Contrast-induced nephropathy was defined as an increase in creatinine by 0.5 mg/dl or >25% of the baseline value.15 Patients were systematically checked for radiation-induced cutaneous injury if the fluoroscopy time exceeded 50 min or if the radiation delivered to the interventional reference point exceeded 4 Gy.16
Complexity classification of the attempted lesions
To classify the attempted CTO lesions according to their complexity, we used the J-CTO classification scheme.9 According to this classification, we assigned every attempted lesion 1 point for each one of the following characteristics: blunt morphology of the entry point of the occlusion, any visible calcification within the occluded segment, tortuosity exceeding 45° within the occluded segment determined by angiography, occlusion length exceeding 20 mm, previous failed attempts at recanalisation. The lesions were classified as easy, intermediate, difficult and very difficult if the total J-CTO score was 0, 1, 2 or ≥3, respectively. Quantitative coronary angiography software (Medis, The Netherlands) was used for accurate measurement of the occlusion length, which was assessed as the distance between the proximal point of the occlusion and the point of distal vessel filling via collateral circulation, as revealed by contralateral injection. When guidewire crossing or retrograde insertion revealed that the true length of occlusion was shorter, the measurement of length was performed in angiographic frames acquired during the procedure.
Statistical analysis
Continuous variables are presented as mean±SD or median (IQR) as appropriate. Categorical variables are presented as absolute numbers and percentages of the total. Comparisons between groups were performed by independent-samples Student t test or by Mann–Whitney U test for continuous variables, and χ2 test for categorical variables. The independent correlates of technical success were assessed in a logistic multivariate model. The candidate explanatory variables for a multivariate model were selected on the basis of significant univariate correlations with the dependent variable. Male gender, hypercholesterolaemia, previous coronary artery bypass graft surgery (CABG), J-CTO score and technical era were used as covariates in the model. Furthermore, aiming to spot differences in the two periods, we assessed separately the univariate associations of technical success with lesion complexity characteristics in each period by simple logistic regression analysis. Preliminary analyses were conducted to ensure that there were no violations of the requirements of each test used. All statistical tests were run with the statistical package, SPSS V.15, and were considered significant at the level of a two-tailed p value <0.05.
Results
Baseline characteristics
Our study population consisted of 483 subjects, aged 63.5±10.4 years, 84.7% of whom were male. Periods 1 and 2 included 288 and 195 patients, respectively. Clinical characteristics of the patients are presented in Table 2. In comparison with period 1, more period 2 patients were hypertensive (92.8% vs 55.6%, p<0.001) and hypercholesterolaemic (92.8% vs 55.6%, p<0.001). A history of myocardial infarction and previous CABG was more common in period 2 than period 1 (70.3% vs 59.0%, p=0.01 and 29.2% vs 14.9%, p<0.001 respectively). In period 2, patients presented with a higher Canadian Cardiovascular Society (CCS) angina class (CCS class III and IV in 72.8% vs 48.3%, p<0.001) and lower left ventricular ejection fraction (52.4±9.4% vs 56.9±10.8%, p<0.001).
Baseline clinical characteristics
Angiographic and procedural characteristics
In period 2, right coronary artery CTOs were attempted more commonly (60% vs 47.2%, p<0.01), and left circumflex artery CTOs less commonly (10.8% vs 19.1%, p=0.01), while left anterior descending artery CTOs were attempted with the same frequency in the two periods. No differences were observed in the presence of CTOs treated within the context of multivessel disease. With regard to the lesion characteristics in period 2, there was a higher incidence of blunt morphology of the entry point (47.2% vs 25%, p<0.001), and the lesions were more often visibly calcified (50.8% vs 34.7%, p<0.001) and tortuous (22.1% vs 9.7%, p<0.001). Moreover, the lesions were longer (22.0 (15.6–32.1) vs 18.2 (12.8–25.4) mm, p<0.001), of longer duration (25 (12–72) vs 11(6–13) months, p<0.001) and had more often been subjected to an unsuccessful attempt at revascularisation (37.9% vs 7.3%, p<0.001). The average J-CTO score was higher in period 2 (1.87±1.2 vs 1.34±1.1, p<0.001), with fewer lesions classified as easy and intermediate (14.9% vs 31.9% (p<0.001) and 18.5% vs 35.4% (p<0.001), respectively) and more lesions classified as difficult and very difficult (30.3% vs 19.8% (p<0.01) and 36.4% vs 12.8% (p<0.001), respectively)(figure 1, table 3). Successful recanalisation was achieved in a higher percentage of cases in period 2, but the difference did not reach statistical significance (85.6% vs 79.5%, p=0.09) (figure 2, table 3).
Angiographic and procedural characteristics
Lesion complexity characteristics in the two periods.
Overall successful recanalisation.
The success rate in easy and intermediate CTOs was similar in period 2 to period 1 (although there was a trend for higher success in period 2). In contrast, in period 2 there was a higher success rate in difficult and very difficult CTOs (88.1% vs 68.4% (p<0.001) and 78.9% vs 42.0% (p<0.001), respectively) (figure 3).
Success per lesion complexity class in the two periods.
In comparison with period 1, the retrograde approach was implemented far more often in period 2 (38.3% vs 10.4%, p<0.001). In addition, the fluoroscopy time was longer by 23 min (p<0.001) in period 2. The median amount of contrast used in period 2 exceeded that used in period 1 by 90 ml (260 ml vs 350 ml, p<0.001). However, no patients were diagnosed with radiation-induced injury throughout the period of our study, and the incidence of contrast-induced nephropathy was low and did not differ in the two periods, with no patients requiring temporary or permanent haemofiltration or dialysis. Similarly, no differences were observed in the incidence of procedure-related myocardial infarction and in major and minor perforations. The in-hospital death occurred in a patient with refractory shock with successful uneventful CTO performed in an attempt to achieve complete multivessel patency 2 years after stenting of the culprit artery late after ST segment elevation myocardial infarction.
Correlates of technical success
The predictors of technical success in the two periods revealed that, in period 1, blunt entry point morphology (OR=0.31, 95% CI 0.17 to 0.58), calcification (OR=0.24, 95% CI 0.13 to 0.45), tortuosity (OR=0.35, 95% CI 0.15 to 0.79) and length of the occluded segment (OR=0.95, 95% CI 0.92 to 0.98) showed a significant reverse relationship with success. In period 2, such an association was present only for occluded segment calcification (OR=0.32, 95% CI 0.13 to 0.81) (table 4).
Univariate associations of lesion characteristics with technical success in the two periods by simple logistic regression analysis
In a logistic regression multivariate model, we assessed the independent correlates of technical success. J-CTO class of lesion complexity (adjusted OR=0.52, 95% CI 0.4 to 0.68), technical era (ie, the availability of sophisticated dedicated devices and materials for CTOs along with the implementation of novel CTO techniques) (adjusted OR=3.3, 95% CI 1.85 to 5.89) and hypercholesterolaemia (adjusted OR=0.39, 95% CI 0.18 to 0.86) were identified as independent correlates of technical success (table 5).
Successful recanalisation correlated factors by multiple logistic regression analysis
Discussion
The primary finding of this study is that, although the absolute success rate of CTO recanalisation has remained relatively stable over a 10-year period, the complexity of the lesions attempted, and treated successfully, has increased dramatically. In addition, many specific lesion characteristics previously associated with procedural failure9 ,17 no longer increase the risk of failure in the modern era because of the availability of novel techniques and improved equipment.
Lesion complexity
The percentage of difficult and very difficult lesions (J-CTO score 2 and ≥3)—aggregating a number of adverse characteristics such as increased occlusion length, calcification, tortuosity, blunt morphology of the entry point, and previously failed attempts—was significantly higher in the second period of our study. During the same period, the successful treatment of these complex lesions became feasible because of advances in recanalisation techniques, materials and dedicated devices. In contrast, easy and intermediate lesions (J-CTO scores 0 and 1) represented a smaller percentage of the lesions attempted in period 2. No differences were observed in the already high success rate in this group of lesions, probably because, in most such cases, the treatment strategy was not significantly different from that in the later period of our study. Moreover, no pronounced improvements would be expected in this already high success rate, and any increases would probably require larger study samples in order to be evident.
Correlates of success
In our total study population, both the J-CTO score of the attempted lesions and the availability of novel techniques and materials were identified as independent determinants of technical success.
As lesions get more complex, technical success is expected to decrease. Difficulties in wire crossing have been previously associated with the anatomical characteristics of the occluded segment such as entry point morphology, bends, calcification and length, as well as with previously failed attempts, which may add to the lesion complexity by creating new anatomical obstacles such as wide dissections or damage to the collateral channels. Retrograde attempts may be hindered by barely connected or tortuous collaterals.
Novel techniques and materials may help overcome, at least partially, the problems arising from adverse lesion characteristics to the extent that recanalisation of occlusions with unfavourable features may be considered feasible. The use of IVUS helps to identify the entry point in blunt stump or stumpless occlusions.4 ,18 The retrograde approach19 offers alternatives for guidewire crossing in occlusions with blunt morphology of the entry point or in stumpless occlusions, while knuckled wire techniques combined with IVUS imaging help negotiate heavily calcified, long or tortuous occlusions. Collateral channel tracking and retrograde crossing of the occlusion has been enhanced with the evolution and refinement of guidewires. Dedicated microcatheters act as collateral channel dilators, allowing enhanced guidewire support and manipulation, crossing collateral channels with retrograde balloons without pre-dilatation, and retrograde wire externalisation.
However, the evolution of CTO technology and techniques in parallel with the complexity of the attempted lesions may provide an explanation for the relative steadiness of the success rates in CTO recanalisation. Indeed, our results are indicative of the volume of complex lesions that are attempted, counterbalancing the increased success rate in this setting, and limiting the potential for increases in the overall success rate.
In line with the above, we have demonstrated the reverse relationship of most of the components of lesion complexity9 with technical success in the early period of our study, but this reverse association was not confirmed in the later period, with the exception of occlusion calcification. A previously failed attempt was not found to correlate with technical failure in either the early or late period. The impact of previous attempts on the success of the procedure is related to complications such as damaging of the collateral channels, thrombosis of the donor vessel, and wide dissections with failure to re-enter the true lumen, the degree of which may vary between various centres and various operators who have performed the previous attempts. In our centre, all CTO recanalisation attempts were performed by the same experienced operator, and special care was taken not to aggravate the patient's condition and not to compromise a future attempt at the same lesion. The avoidance of very aggressive treatment techniques, involving wide dissections (Sub-intimal Tracking And Re-entry), along with the practical difficulties imposed by the calcium burden on crossing the occluded segment, in the non-negligible number of cases where the antegrade approach was implemented in period 2, may explain the persistence of the correlation of lesion calcification with technical failure throughout the time span of our study. It is worth mentioning that the above associations were investigated toward the end point of procedural success irrespective of the procedural time. In concordance with previous reports,9 a longer procedural time was required for recanalisation of highly complex lesions, as reflected in the longer fluoroscopy time, and the use of contrast was increased in the later period of our study.
Patient characteristics
In tandem with the complexity features seen in lesions treated in the late period, the patients’ profiles also changed, with traditional risk factors for cardiovascular disease, such as hypertension and hypercholesterolaemia, along with a history of myocardial infarction and previous CABG, appearing more frequently. Moreover, the patients presented with more severe angina and with a lower left ventricular ejection fraction. These features, despite not having an obvious effect on the technical success of the procedure, may imply the treatment of higher-risk patients who would benefit most from revascularisation. In our study, hypercholesterolaemia was identified as an independent correlate of successful recanalisation. We hypothesise that this association is based on the effect of high cholesterol on the anatomical and/or histopathological characteristics of the lesion. In contrast with previous studies, previous CABG was not found to correlate with failure, and this may be explained by the use of grafts for successful retrograde recanalisation in several cases in our study.
Safety
The increased rate of attempting and recanalising more complex lesions was achieved with an acceptable safety profile, as it was not accompanied by any increases in procedural and periprocedural complications. The only post-procedural in-hospital death occurred in a patient with refractory shock in whom a successful uncomplicated recanalisation of a chronic right coronary artery occlusion was achieved as part of a strategy of multivessel treatment extended to the non-culprit arteries. Procedure-related myocardial infarction,20 major and minor coronary artery perforations,21 and contrast-induced nephropathy22 occurred with the same frequency through both periods of our study and in low percentages comparable with those in standard PCIs in patients with similar clinical characteristics. No patient required temporary or permanent haemodialysis and no cases of radiation-induced cutaneous injury occurred in our study population.
Study limitations
A single-operator study is certainly not an ideal model for generalising results. However, the influence of new operators in a practice, as well as the technical progress of young operators still on the steep slope of their learning curve, represents a confounding variable, which could invalidate results. A single-centre, single-operator study is unlikely to represent a limitation when the operator already had more than 15 years of PCI practice with 250–400 procedures per year. The observational nature of our study limits the potential for inferences regarding cause-and-effect relationships between lesion complexity characteristics, techniques and equipment and successful CTO recanalisation. However, it is conceivable that changes in CTO-specific technologies would affect either the recanalisation success rate or the complexity of the CTOs successfully treated. The size of our database, which was built with consecutive CTO cases, thereby excluding any selection bias, adds to the robustness of our observations.
Clinical implications
CTO recanalisation represents one of the most challenging and technically demanding aspects of percutaneous treatment of coronary artery disease. The lower success rate in percutaneous treatment of CTOs in comparison with conventional PCI, which has been relatively steady despite the advances in CTO-specific techniques and equipment, represents a major factor for selecting other management strategies such as surgical treatment or conservative medical therapy. However, our study provides insights into the aetiology of this seemingly faultless success, according to which the enhanced potential offered by the novel approaches and technology in CTO treatment is balanced by the increased complexity of the lesions attempted. The true message of these findings is that, in experienced hands, CTO recanalisation is feasible and relatively safe, even in patients with high-complexity lesions and clinical characteristics denoting higher risk, for whom more benefits would be expected from such a procedure.
Conclusions
In summary, evolution in techniques and specific technologies for percutaneous CTO recanalisation has dramatically increased the rate of successful treatment of occlusions with high-complexity features. The increase in successful treatment in this setting has been counterbalanced by the increase in the percentage of such complex lesions attempted. This may explain the stability of total success rates despite the relevant technical advances and improvements in equipment. However, percutaneous treatment of CTOs previously considered more suitable for surgical or medical treatment now seems feasible and with acceptable risk.
References
Footnotes
-
Contributors DS: planning, conduct, and reporting. CD: planning, conduct, and reporting. AL: planning, and reporting. MG: conduct and reporting. KK: conduct and reporting. NF: planning. EAB: conduct. GGS: conduct. RM: conduct. All authors have given approval for the final version of the manuscript to be published.
-
Competing interests None.
-
Patient consent Obtained.
-
Provenance and peer review Not commissioned; externally peer reviewed.