Impact of Drug-Coated Balloon-Based Revascularization in Patients with Chronic Total Occlusions

Background: Percutaneous coronary intervention (PCI) with a drug-eluting stent (DES) for chronic total coronary occlusions (CTOs) improves clinical symptoms and quality of life. However, data on drug-coated balloon (DCB)-based PCI in CTO lesions are limited. Methods: A total of 200 patients were successfully treated for CTO lesions, either with DCB alone or in combination with DES (DCB-based PCI). They were compared with 661 patients who underwent second-generation DES implantation for CTO from the PTRG-DES registry (DES-only PCI). The endpoint was major adverse cardiovascular events (MACEs), which included a composite of cardiac death, myocardial infarction, stent or target lesion thrombosis, target vessel revascularization, and major bleeding at 2 years. Results: In the DCB-based PCI group, 49.0% of patients were treated with DCB only and 51.0% underwent the hybrid approach combining DCB with DES. Bailout stenting was performed in seven patients (3.5%). The DCB-based PCI group exhibited fewer stents (1.0; IQR: 0.0–1.0 and 2.0; IQR: 1.0–3.0, p < 0.001), shorter stent lengths (6.5 mm; IQR: 0.0–38.0 mm and 42.0 mm; IQR: 28.0–67.0 mm, p < 0.001), and lower usage of small stents with a diameter of 2.5 mm or less (9.8% and 36.5%, p < 0.001). Moreover, the DCB-based PCI group had a lower rate of MACEs than the DES-only PCI group (3.1% and 13.2%, p = 0.001) at 2-year follow-up. Conclusions: The DCB-based PCI approach significantly reduced the stent burden, particularly in the usage of small stent diameters, and resulted in a lower risk of MACEs compared to DES-only PCI in CTO lesions.


Introduction
Chronic total occlusion (CTO) of a coronary artery is relatively common, occurring in approximately 16% of patients diagnosed with coronary artery disease (CAD) [1].Recent advancements in devices and treatment algorithms and growing experience have led to increased technical success in CTO percutaneous coronary intervention (PCI); however, the results regarding the clinical benefit of CTO remain inconclusive [2][3][4][5].
CTO PCI typically requires the use of multiple lengthy stents due to the lesion's characteristics, leading to an increase in stent length.In the DECISION-CTO trial, an average of 2.3 stents were used, with a total length of 68.2 mm [4].It is well known that a longer stent length independently predicts in-stent restenosis and stent thrombosis [6][7][8][9].Furthermore, "full-metal jacket" PCI using overlapping drug-eluting stent (DES) is associated with a high rate of adverse events [10].Moreover, stent implantation can hinder the restoration of vasomotion in stented segments and accelerate the development of neoatherosclerosis [11][12][13].
The use of drug-coated balloons (DCBs) in treating patients with in-stent restenosis [14] and de novo small vessel disease [15,16] has demonstrated comparable clinical outcomes to DES implantation.Unlike DES implantation, the DCB strategy leaves no material behind, thereby reducing the risk of stent-related adverse biological reactions that can lead to restenosis and thrombosis [17].This approach promotes vascular healing.In particular, using DCB either alone or in combination with DES as a part of a hybrid procedure to decrease stent burden, especially in cases of small diameter stent implantation, could offer an alternative and beneficial treatment approach for CTO lesions.Recently, we reported that the DCB-based PCI approach, guided by the International Consensus Group [18] and Asian Pacific Consensus Group [19], significantly reduced stent burden for multivessel PCI, resulting in a lower rate of major adverse cardiovascular events (MACEs) at 2 years compared to DES-only PCI [20].This study demonstrated the safe reduction in secondgeneration DES burden for lesions requiring multiple stents using the DCB-based PCI approach.However, the efficacy of DCB-based PCI for patients with CTO CAD has not been fully validated in the modern DES era.Hence, our aim was to evaluate the clinical significance of DCB-based PCI in patients with CTO CAD who underwent PCI with secondgeneration DES.

Patient Population
A retrospective enrollment included a total of 200 patients who underwent successful PCI for CTO CAD using DCB alone or in combination with the DES hybrid approach.The patients were recruited from three university hospitals in South Korea (Ulsan University Hospital, Kangwon National University Hospital, and Korea University Ansan Hospital), all of which have experience in treating patients with CTO CAD using DCB (Impact of Drug-coated Balloon Treatment in de Novo Coronary Lesion; NCT04619277).
We excluded patients meeting any of the following criteria from our analysis: a history of coronary artery bypass surgery, presentation with cardiogenic shock, administration of thrombolysis prior to PCI, suboptimal or failed PCI for target lesions, or loss to follow-up.Subsequently, we compared these findings with data from patients exclusively treated with DES for CTO CAD within the PTRG-DES (Platelet function and genoType-Related long-term ProGnosis in Drug-Eluting Stent-treated patients with coronary artery disease) consortium (ClinicalTrials.govIdentifier: NCT04734028) [21].The study protocol received approval from the institutional review board of each participating center (approval code: 2021-07-036; approval date: 18 August 2021), and all patients provided written informed consent at the time of enrollment.

Procedure
All patients received 200 mg aspirin and 300 or 600 mg clopidogrel as loading doses before the procedure.Before proceeding with diagnostic coronary angiography, intracoronary nitroglycerin (200 µg) was administered to the patients.In CTO lesions, predilation balloon angioplasty was carried out to assess the feasibility of DCB treatment.For the DCB-based treatment group, interventions followed the recommendations outlined in international and Asia-Pacific consensus guidelines for DCB treatment [18,19].All patients underwent predilation with either semi-compliant, non-compliant, or scoring balloons, adhering to the recommended balloon-to-vessel ratio of 0.8 to 1.0.Following predilation balloon angioplasty, stenting was postponed in cases of all types of dissections, as long as thrombolysis in myocardial infarction (TIMI) flow grade 3 was present.However, if predilation resulted in flow-limiting dissection (TIMI flow grade < 3), stent implantation was recommended without using a DCB.The DCB was inflated to its nominal pressure for a minimum of 60 s, ensuring that it extended at least 2 mm beyond the length of predilatation.Each DCB was coated with 3.0 µg/mm 2 of paclitaxel combined with iopromide (SeQuent Please, B. Braun) as a carrier for the drug.Following the use of DCB, the final assessment was performed at least 5 min after administering a bolus of intracoronary vasodilator to prevent potential acute vessel closure.In cases with a high thrombus burden, a rescue strategy involving the use of glycoprotein IIb/IIIa receptor inhibitors was employed.The duration of the dual antiplatelet therapy was determined at the discretion of the attending physician.

Clinical Follow-Up and Endpoints
All 861 patients underwent clinical follow-up after the index procedure through outpatient clinic visits and telephone interviews.The study endpoint was the cumulative occurrence of MACEs, which included cardiac death, myocardial infarction, stroke, stent or target lesion thrombosis, target vessel revascularization, and major bleeding within a 2-year period.Cardiac death was defined as any death not distinctly attributable to non-cardiac causes, including myocardial infarction, according to previously published guidelines [11].The definition of definite stent thrombosis adhered to the parameters outlined by the Academic Research Consortium definition [22], and major bleeding was defined as Bleeding Academic Research Consortium types 3 to 5 bleeding [23].

Statistical Analysis
Clinical characteristics were presented as percentages for categorical variables and as means with standard deviations or medians with interquartile ranges (Q1-Q3) for continuous variables.Group comparisons were performed using Pearson's Chi-squared test or Fisher's exact test for categorical variables and Student's t-test or Mann-Whitney Utest for continuous variables, as appropriate.To compare clinical outcomes between the two groups, we evaluated the cumulative incidences of MACEs and other outcomes using the Kaplan-Meier method and then compared the curves using the log-rank test.To minimize the potential influence of confounding factors, we employed propensity score matching, inverse probability weighted (IPW) Cox proportional hazard regression, and multivariable Cox proportional hazard regression to adjust for differences in baseline characteristics.The propensity score was calculated through logistic regression taking into account clinical and lesion variables (age, sex, hypertension, diabetes mellitus, current smoking, previous MI, previous PCI, end-stage renal disease, clinical presentation status, left ventricular ejection fraction, left main disease, bifurcation lesion, and the number of diseased vessels).A 1:1 nearest neighbor matching without replacement was performed with a caliper size of 0.2 of the propensity score.Following propensity score matching or IPW adjustment, we evaluated the balance between the two groups by calculating absolute standardized mean differences.Successful balance between the comparative groups was considered as achieved when absolute standardized mean differences for all relevant covariates were 0.1 or less (Supplementary Table S1).
Survival curves for 2-year MACEs were constructed using Kaplan-Meier estimates and compared using the log-rank test.Hazard ratios (HRs) with 95% confidence intervals (CIs) for the treatment groups as an independent predictor of MACEs were calculated through multivariable Cox regression analysis.The assumption of proportionality was assessed using two-sided p-values, and a value of <0.05 was considered statistically significant.Statistical analyses in this study were conducted using R version 4.3.2(R Foundation for Statistical Computing, Vienna, Austria).

Results
From October 2010 to November 2022, 1388 patients underwent PCI based on DCB, of which 53 patients had coronary artery bypass surgery, while 1135 patients had no CTO lesions; therefore, 200 patients were included in the analysis (Figure S1).In the DES treatment group, a total of 13,160 patients were screened, with DES treated between July 2003 and August 2018, which was from the PTRG-DES consortium, of which 11,226 patients utilized second-generation DESs.Excluding 139 patients who had previously undergone coronary artery bypass graft surgery, we further excluded 10,329 patients without CTOs.Supplementary Figure S1 illustrates the study group diagram, indicating how the final population included in this study was identified.
The baseline clinical and lesion characteristics of the patients are summarized in Tables 1 and 2. A total of 861 patients underwent PCI successfully for CTO lesions.The mean age of the patients was 63.3 ± 11.1 years, and the majority of patients were men.The two groups differed significantly in clinical and lesion characteristics across the overall population.The DCB treatment group had a higher proportion of men, previous PCI, and more frequent end-stage renal disease, presentations with unstable angina, left main disease, bifurcation disease, and target lesion of left circumflex artery.The DES-only group had a higher prevalence of dyslipidemia, current smokers, and acute MI presentation.In the DCB-based group, 49.0% of the patients were treated with DCB alone and 51.0% were treated with the hybrid approach using DESs.For the devices used in the DCB-based group, 66.0% were DCBs and 34.0% were DESs.In the DCB-based group, bailout stenting was performed in seven patients (3.5%).The procedure characteristics are summarized in Table 2.The total number of diseased vessels, the total number of treated vessels, and the In the DCB-based group, 49.0% of the patients were treated with DCB alone and 51.0%In the DCB-based group, 49.0% of the patients were treated with DCB alone and 51.0% were treated with the hybrid approach using DESs.For the devices used in the DCB-based group, 66.0% were DCBs and 34.0% were DESs.In the DCB-based group, bailout stenting was performed in seven patients (3.5%).The procedure characteristics are summarized in Table 2.The total number of diseased vessels, the total number of treated vessels, and the total number of used devices were higher in the DCB-based PCI group.The total device length was higher in the DCB-based group (60.0 mm; IQR: 36.5-86.0mm in the DCB-based group; and 42.0 mm; IQR: 28.0-67.0mm in the DES-only group, p < 0.001), but the mean device diameter was larger in the DES-only group (2.8 mm; IQR: 2.5-3.0 mm) compared to that of the DCB-based group (2.7 mm; IQR:2.5-2.9 mm) (p < 0.001).
A statistically significant difference was observed in the number of DCBs and DESs used in both the groups.In the DCB-based group, the number of DCBs was 1.0 (IQR: 1.0-2.0)and that of DESs was 1.0 (IQR: 0.0-1.0).The total DCB length in the DCB-based group was 30.0 mm (IQR: 26.0-56.0mm), and the mean DCB diameter was 2.5 mm (IQR: 2.5-2.7 mm).In the DCB-based group, a small DCB with a diameter of ≤2.5 mm was utilized in 64.5% of the patients.There were significant differences observed in the total lengths and mean diameters of the DESs used between the two groups.The total DES length was 6.5 mm (IQR: 0.0-38.0mm) in the DCB-based group and 42.0 mm (IQR: 28.0-67.0mm) in the DESonly group (p < 0.001).The mean DES diameter was larger in the DCB-based group and was 3.0 mm (IQR: 2.8-3.5 mm) compared to 2.8 mm (IQR: 2.5-3.0 mm) in the DES-only group (p < 0.001).A significant disparity between the two groups was evident when the DES diameter was ≤2.5 mm (9.8% in the DCB-based group vs. 36.5% in the DES-only group, p < 0.001).In comparison with the DES-only group, the DCB-based group saw a notable reduction in both the number of stents and the total stent length by 66.0% and 63.9%, respectively, in the DCB-based group compared with those of the DES-only group.

Two-Year Outcomes
Clinical follow-up was available in all patients (100%).A comparison of clinical outcomes between the DCB-based PCI group and the DES-only PCI group is presented in Table 3 and Figure 1.The median follow-up duration was 1155.0 days (IQR: 424.5-2004.0days) in the DCB-based group and 466.0 days (IQR: 365.0-1653.0days) in the DES-only PCI group.The risk of MACEs was significantly lower in the DCB-based PCI group than that in the DES-only PCI group (3.1% vs. 13.2%;HR: 0.21; 95% CI: 0.08-0.51;p = 0.001).The risk of target vessel revascularization and major bleeding was also significantly lower in the DCB-based PCI group than that in the DES-only PCI group (1.9% vs. 7.3%; HR: 0.24; 95% CI: 0.08-0.79;p = 0.019).Major bleeding was higher in the DES-only group (0 vs. 4.6%; HR: 0.06; 95% CI: 0.00-0.41;p = 0.001).No target lesion thrombosis and major bleeding were observed in the DCB-based treatment group.Representative images are shown in Figure 2.
Across sensitivity analyses utilizing multivariable Cox regression, propensity score matching and IPW adjustment consistently lower the risks of MACEs in the DCB-based PCI group compared to those in the DES-only PCI group (Table 3).

Independent Predictors of Composite Outcomes
Multivariable Cox proportional hazard models identified the independent predictors of MACEs and target vessel revascularization (Supplementary Table S2).DCB-based PCI was independently associated with a decreased risk of MACE (HR: 0.21; 95% CI: 0.08 to 0.52; p = 0.001) and target vessel revascularization (HR: 0.25; 95% CI: 0.07 to 0.81; p = 0.021) at 2 years.were observed in the DCB-based treatment group.Representative images are shown in Figure 2.

Discussion
The main finding of the present study was that the DCB-based treatment approach for CTO CAD, either using DCB alone or in combination with DES guided by the Interna-

Discussion
The main finding of the present study was that the DCB-based treatment approach for CTO CAD, either using DCB alone or in combination with DES guided by the International Consensus Group [18] and Asian Pacific Consensus Group [19], significantly and safely reduced the stent length and the number of stents.This reduction was particularly notable in terms of the small diameter of DES.This resulted in a lower risk of MACEs at 2 years compared to that of the DES-only treatment.Furthermore, sensitivity analyses using multivariable Cox regression, propensity score matching, and IPW adjustment consistently showed significantly lower risks of MACEs in the DCB-based PCI compared to those of the DES-only PCI group.Therefore, our data suggest that the DCB-based treatment approach is a reasonable strategy to improve outcomes in patients with CTO CAD.
In a society with an increasingly aging population, the issue of CTO CAD has become a challenge for interventional cardiologists.The CTO lesion is not rare, and a big registry reported that 16% of patients with CAD had CTO lesion [1].Thus far, despite advancements in CTO recanalization techniques, operator expertise, a more standardized approach, and improved devices, including DES platforms, the results regarding the clinical advantages of CTO PCI over medication alone remains inconclusive [2][3][4]24].Given the nature of the lesion, CTO often requires the use of a lengthy stent, similar to treating a diffuse long lesion.A long stent length is known to independently predict in-stent restenosis and stent thrombosis [6][7][8].Recently, the GRAND-DES registry with the use of second-generation DESs demonstrated that long stent length (>40 mm) was significantly associated with higher target lesion failure and definite or probable stent thrombosis [9].The study also demonstrated that long stent length is an independent predictor for ischemic clinical events in patients undergoing PCI [8].Furthermore, performing a "full-metal jacket" PCI using overlapping DESs is associated with a high adverse event rate [10].In a previous trial, two independent predictors of target lesion failure in "full-metal jacket" PCI were identified: number of target vessel DESs (hazard ratio: 1.72; 95% confidence interval: 1.16 to 2.54; p = 0.006) and persistent distal luminal narrowing (hazard ratio: 2.73, 95% confidence interval: 1.66 to 4.47; p < 0.001) [25].Therefore, many CTO operators recommend avoiding the overtreatment of distal lesions with DESs because they often improve over time after the restoration of antegrade flow [26][27][28].Additionally, stent implantation can hinder the restoration of vasomotion in stented segments and accelerate the development of neoatherosclerosis [11][12][13].
The use of modern, new-generation DESs has shown significant improvements in reducing adverse events compared to bare-metal stents and first-generation DESs.Nevertheless, contemporary DESs still pose risks of short-and long-term device-related adverse cardiovascular events due to their permanent metallic cage, which can increase the likelihood of stent thrombosis and in-stent restenosis [29][30][31].CTO CAD, particularly involving complex and long lesions in small vessels, presents a challenging scenario in interventional cardiology with no unequivocal evidence on the best treatment approach.To address these concerns, efforts have been made to develop new treatment strategies, such as the hybrid approach involving a combination of DCBs and DESs.The hybrid approach offers the advantage of reducing the length of the permanent metallic cage, potentially leading to lower rates of stent thrombosis and restenosis.In this strategy, consecutive lesions are addressed individually, where a DES is placed in the proximal segment, including the CTO entrance, and a DCB treatment is performed for the distal lesion, which experiences negative remodeling due to the absence of antegrade flow for long time.Several observational studies have shown the feasibility and safety of this approach [32,33].Similar to the previous study, this study also reduced the stent length, including the number of stents, in the DCB-based treatment group, with a particular focus on reducing the utilization of small stent diameters.This approach was employed for over half of the patients.In these cases, DESs were utilized in the larger, more proximal lesions, such as CTO entrance sites, while DCBs were reserved for smaller, more distal coronary lesions.This hybrid approach offers several advantages, including minimizing the overall stent burden, especially in distal le-sions with smaller diameters, and treating the underlying disease while maintaining access for potential future coronary artery bypass grafting, if necessary.Decreasing metal stent length may also help maintain the vessel's response to vasomotion stimuli and reduce the risk of neoatherosclerosis [17].While the development of devices, procedural techniques, and medication regimens to prevent very late events is crucial for improving lifelong outcomes in patients undergoing coronary revascularization, adopting a DCB-based treatment approach offers potential benefits for managing CTO lesions.Further studies are needed for a comprehensive evaluation of the role of DCB in this context.

Limitations
Our study has some limitations.Firstly, it is crucial to acknowledge that this study possesses inherent limitations stemming from its observational nature and reliance on registry data.Furthermore, the allowance for physicians to opt for the treatment strategy introduces the potential for selection bias.To mitigate this bias, extensive sensitivity analyses were undertaken, adjusting for both measured and unmeasured confounders to minimize the impact of varying baseline characteristics.Secondly, the study population was drawn from an expert center specialized in exclusive DCB treatment for de novo CAD.Consequently, the findings might not be easily replicable without a substantial learning curve and proficiency in this particular treatment approach.Thirdly, discrepancies in the results might arise due to disparities in the enrollment periods of the two groups.While the PTRG-DES registry has been operational since 2003, the patients included in the propensity match analysis were instances involving second-generation DESs.Hence, significant differences stemming from device development and enhancements in PCI techniques between the two groups are not anticipated.For more robust and comprehensive insights into the long-term outcomes of DCB-based treatment in CTO lesion PCI, further prospective randomized non-inferiority or superiority clinical trials with more extensive patient cohorts are imperative.

Conclusions
Despite improvements in modern DESs, concerns remain about negative effects related to stents, especially for patients with CTO CAD.Using the DCB-based treatment approach, either with DCB alone or combined with DESs, has shown great potential in reducing the stent-related adverse events, particularly by reducing the use of smaller DESs.Additionally, this approach led to a lower risk of MACEs one year after treatment compared to that when only DESs were used.Our study adds to the growing evidence that supports the DCBbased approach as a promising strategy for effectively managing this difficult condition.To further confirm the safety and effectiveness of the DCB-based treatment in CTO PCI, future research, including larger randomized controlled trials, will be crucial.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jcm13123381/s1,Table S1.Absolute standardized differences of variables among unadjusted, propensity-score matched, and IPW-adjusted cohort.Table S2.Independent predictors for major adverse cardiovascular events or target vessel revascularization.

Figure 3
Figure 3 presents the prognostic impact of DCB-based PCI among various subgroups.The significantly reduced risk of MACEs in the DCB-based PCI group compared to that in the DES-only PCI group remained consistent across all subgroups with no significant interaction p values.

Figure 1 .
Figure 1.Cumulative incidence of major adverse cardiovascular events and target vessel revascularization in the overall population and propensity-matched population.(A) Major adverse cardiovascular events in overall population.(B) Target vessel revascularization in overall population.(C) Major adverse cardiovascular events in propensity-matched population.(D) Target vessel revascularization in propensity-matched population.DCB, drug-coated balloon; DES, drug-eluting stent; PCI, percutaneous coronary intervention; HR, hazard ratio; and CI, confidence interval.

Figure 1 .
Figure 1.Cumulative incidence of major adverse cardiovascular events and target vessel revascularization in the overall population and propensity-matched population.(A) Major adverse cardiovascular events in overall population.(B) Target vessel revascularization in overall population.(C) Major adverse cardiovascular events in propensity-matched population.(D) Target vessel revascularization in propensity-matched population.DCB, drug-coated balloon; DES, drug-eluting stent; PCI, percutaneous coronary intervention; HR, hazard ratio; and CI, confidence interval.

Figure 1 .
Figure 1.Cumulative incidence of major adverse cardiovascular events and target vessel revascularization in the overall population and propensity-matched population.(A) Major adverse cardiovascular events in overall population.(B) Target vessel revascularization in overall population.(C) Major adverse cardiovascular events in propensity-matched population.(D) Target vessel revascularization in propensity-matched population.DCB, drug-coated balloon; DES, drug-eluting stent; PCI, percutaneous coronary intervention; HR, hazard ratio; and CI, confidence interval.

Figure 2 .Figure 2 .
Figure 2. Representative case from the drug-coated balloon-based treatment group.(A) A CTO lesion in the LAD.The arrow points to the CTO entrance.(B) Angiogram after predilation.(C) A DES Figure 2. Representative case from the drug-coated balloon-based treatment group.(A) A CTO lesion in the LAD.The arrow points to the CTO entrance.(B) Angiogram after predilation.(C) A DES implantation at proximal LAD.(D) After DES implantation, diffuse stenotic lesion is seen in the middle of LAD.(E) It was treated with a small diameter DCB.(F) Follow-up angiogram after 6 months.CTO, chronic total occlusion; LAD, left anterior descending artery; DES, drug-eluting stent; and DCB, drug-coated balloon.

Figure S1 .
Study group diagram indicating how the final population included in this study was identified.DCB indicates drug-coated balloon; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft; CTO, chronic total occlusion; DES, drug-eluting stent; LVEF, left ventricular ejection fraction.Author Contributions: E.-S.S. and H.B.L. designed this study and contributed to the final analyses and data interpretation.This report was drafted by E.-S.S. and M.H.J.All authors approved the final version of the manuscript and ensured that the accuracy or integrity of any part of the work was appropriately investigated and resolved.All authors accessed and E.-S.S., M.H.J., B.K., S.K., A.-Y.H. and H.B.L. verified the data in the study and were assigned the final responsibility for the decision to submit the work for publication.All authors have read and agreed to the published version of the manuscript.

Table 1 .
Baseline clinical characteristics in the overall population and the propensity-matched population.

Table 3 .
Comparison of 2-year clinical outcomes according to treatment strategy.
Values are n (%) unless otherwise indicated.The cumulative incidences of clinical outcomes are presented as Kaplan-Meier estimates.Multivariable Cox proportional hazard regression model, propensity score-matched cohort, and inverse probability of treatment weighting method were used to adjust for baseline differences between comparative groups.Major adverse cardiovascular events were composed of cardiac death, myocardial infarction, stent or target lesion thrombosis, target vessel revascularization, and major bleeding (Bleeding Academic Research Consortium bleeding type 3 or greater).DCB, drug-coated balloon; DES, drug-eluting stent; PCI, percutaneous coronary intervention; HR, hazard ratio; CI, confidence interval; and IPW, inverse probability weighted.J. Clin.Med.2024, 13, x FOR PEER REVIEW 6 of 12