Efficacy and long-term outcomes of drug coated balloon in de novo lesions of small versus large coronary vessels

Abstract

Objective

Drug eluting stent (DES) could result in both in-stent restenosis and high bleeding risk due to long-term anti-platelet therapy. Drug-coated balloon (DCB) delivers anti-proliferative drugs without implanting metal into vascular wall. Our aim was to investigate its feasibility in large vessel coronary artery disease (LvCAD), compared to small vessel coronary artery disease (SvCAD).

Methods

This study enrolled 237 patients with de novo coronary lesions treated with DCB-only strategy and categorized according to the reference vessel diameter of 3 mm into SvCAD and LvCAD groups. The primary endpoint was in-lesion late lumen loss (LLL). The secondary endpoints included composite major adverse cardiac events (MACE), cardiac death, non-fatal myocardial infarction (MI), target lesion revascularization (TLR), target vessel revascularization (TVR), and vessel thrombosis.

Results

The immediate (3.06 ± 0.25 vs. 2.33 ± 0.21 mm, p = 0.001) and follow up minimal lumen diameter (3.13 ± 0.25 vs. 2.41 ± 0.21 mm, p = 0.001) and acute gain (1.92 ± 0.29 vs. 1.5 ± 0.26 mm, p = 0.04) were significantly higher in LvCAD group. In-lesion LLL was negative without significant difference (-0.07 ± 0.02 vs. - 0.06 ± 0.04 mm, p = 0.69). The incidence of adverse clinical events was not statistically significant accounting for 6.5 % vs. 10.5 % for composite MACE (p = 0.27), 0.8 % vs. 0.9 % for cardiac death (p = 0.96), 4.9 % vs.7 % for non-fatal MI (p = 0.49), 4.1 % vs. 6.1 % for TLR (p = 0.47), 2.4 % vs. 3.5 % for TVR (p = 0.63) and 1.6 % vs. 2.6 % for vessel thrombosis (p = 0.59).

Conclusion

DCB-only strategy is effective in treating LvCAD with comparable outcomes to SvCAD.

1. Introduction

To date, percutaneous coronary intervention (PCI) with implanting drug eluting stent (DES) is still the mainstay non-pharmacological approach for treating patients with coronary artery disease (CAD).¹ However, these metallic implants could be associated with chronic inflammatory response within vascular wall with subsequent development of neoatherosclerosis, restenosis, and late instent thrombosis.² In addition, other complications resulting from DES implantation such as high bleeding risk attributed to long-term use of double anti-platelet therapy (DAPT), stent fracture, and allergy to its polymer have to be seriously considered.³ Thus, It is valuable to use an interventional approach to dilate and treat coronary lesions without permanent implantation of metallic devices.

Drug-coated balloon (DCB) is a novel interventional modality used during PCI by rapidly delivering anti-proliferative drugs into the vascular wall through a specified matrix without implanting permanent metal. Accordingly, the possible complications following DES implantation; especially restenosis and stent thrombosis, could be dramatically reduced due to anti-proliferative nature of the drug covering DCB and its quick transfer to vascular wall.⁴ Consequently, the use of DCB modality in PCI could represent a cutting-edge substitute for DE due to its relatively simple technique, reduced amount of used contrast, shorter DAPT duration up to 1–3 months and absence of metal and polymer in the coronary vessel wall.

Currently, the efficacy of DCB has been already documented in in-stent restenosis (ISR) and small vessel CAD (SvCAD).⁵ Numerous studies concerning the use of DCB to treat de novo coronary lesions have documented that the resulting immediate and long-term outcomes are non-inferior to those following DES implantation.⁶ The BASKET SMALL trial enrolled patients with SvCAD < 3 mm in diameter and reported that DCB angioplasty was non-inferior to new generation DES angioplasty in terms of major adverse cardiac events (MACE) (7.5 % vs. 7.3 %, p = 0.918).⁷ Furthermore, the meta-analysis conducted by Megaly and co-workers confirmed the same conclusion that DCB angioplasty of small sized de novo coronary lesions had comparable target lesion revascularization (TLR) rates in comparison to large vessel (LvCAD) treated with DES.⁸ Nevertheless, these studies were mainly limited to SvCAD, with only few insufficient data promoting a DCB-only strategy for treating denovo LvCAD ≥2.8 mm in diameter.

Recently, the use of DCB to treat patients with LvCAD has gained much popularity in recent studies. This is because treating SvCAD with DCB accounts only for approximately 35 % of all PCI.⁹ Yu et al conducted a prospective trial to assess the clinical effectiveness of DCB in treating patients with LvCAD, and comparing that with those having SvCAD. There was no significant difference regarding long term prognosis between patients with LvCAD and those with SvCAD, demonstrating the safety and efficacy of the DCB-only strategy in treating sizable coronary lesions.¹⁰ However, these findings need bigger sample size and longer follow up period to verify its clinical implication.

In the present study, we retrospectively enrolled patients who treated with DCB-alone strategy for de novo coronary lesions aiming to investigate its feasibility and safety in treating patients with LvCAD, and comparing the resulting outcomes with those resulting from SvCAD.

2. Methods

2.1. Study design

This retrospective cohort study enrolled 260 consecutive patients with de novo coronary lesions treated with DCB-only strategy in our institution from January 2022 to October 2023. The inclusion criteria included patients having de novo coronary lesions ≥ 70 % angiographic stenosis treated with DCB-only strategy. The DCB used in this study had a paclitaxel-iopromid matrix coating (SeQuent® Please, B. Braun Melsungen AG). Twenty three patients were excluded from this study due to contraindications to dual antiplatelet therapy (DAPT), concomitant DES implantation, angioplasty of ISR, loss of follow up, and lack of informed consent.

Finally, 237 patients were included in this study, and categorized according to the reference vessel diameter (RVD) of 3 mm into SvCAD group (123 patients) with RVD ≤3 mm and LvCAD group (114 patients) with RVD >3 mm (Fig. 1). All patients underwent one year clinical and angiographic follow up. The study was conducted in adherence to the principles of Helsinki Declaration. The institutional Review Board of our institution approved the study protocol. All enrolled patients gave written informed consent before the procedure.

Fig. 1.

The enrolment process.

This study enrolled 260 patients with de novo coronary lesions ≥ 70 % stenosis treated with DCB-only strategy. Twenty three patients were excluded due to contraindications to DAPT, concomitant DES implantation, angioplasty of ISR, loss of follow up, and lack of informed consent. Finally, 237 patients were included and categorized according to RVD of 3 mm into SvCAD group (123 patients) with RVD ≤ 3 mm and LvCAD group (114 patients) with RVD > 3 mm. All patients underwent one year clinical and angiographic follow up.

DAPT: dual anti-platelet therapy; DCB: drug coated balloon; DES: drug eluting stent; ISR: in-stent restenosis; LvCAD: large vessel coronary artery disease; LLL: late lumen loss; MACE: major adverse cardiac event; MI: myocardial infarction; RVD: reference vessel diameter; SvCAD: small vessel coronary artery disease; TLR: target lesion revascularization; TVR: target vessel revascularization.

2.2. Data collection

Data were collected from the medical records of our institution including demographic, clinical, laboratory, and procedural characteristics. Cardiovascular risk factors including hypertension, diabetes mellitus, dyslipidemia, chronic kidney disease (CKD), and family history of CAD were identified based on current standard guidelines.¹¹ Resting ECG and comprehensive echocardiographic data including diastolic functions and left ventricular ejection fraction (LVEF) were obtained from all patients before coronary angioplasty. Venous blood samples were withdrawn prior to coronary angiography with reporting of laboratory parameters. Lesion and procedural characteristics regarding target vessels and its number, procedural time, fluoroscopy time, and contrast amount were also documented.

2.3. Coronary angiography protocol

All patients were given an initial heparin bolus dose of 70–100 IU/kg with an extra dose of 1000 IU per hour during the procedure. Loading DAPT dose was given (300 mg aspirin & 300 mg clopidogrel or 180 mg ticagrelor) prior to coronary angioplasty. The radial or femoral arterial access was used with administration of intracoronary 100–200 μg nitroglycerin. At least two orthogonal angiographic views were obtained for better assessment of culprit vessel. Glycoprotein IIb/IIIa antagonists were administered according to operator's decision.

2.4. Quantitative coronary angiography (QCA)

All culprit coronary lesions were analysed using built-in QCA software by experienced interventional cardiologists. The following measurements were recorded: RVD, minimal lumen diameter (MLD), lesion length, percentage of diameter stenosis, and acute gain. In addition, DCB length, diameter, inflation pressure and duration were recorded. These measurements were obtained in triplicate with the calculation of the mean value.

2.5. DCB angioplasty protocol

Before DCB inflation, lesion preparation was done using normal, cutting, and non-compliant balloons with balloon/vessel diameter ratio of 0.8–1.0 to avoid intimal dissection prior to DCB inflation. Satisfactory pre-dilation results were defined as residual stenosis less than 30 %, thrombolysis in myocardial infarction (TIMI) grade 3 flow, and no or type A/B dissections based on NHLBI65 criteria 8. After that, the patient could be considered eligible for DCB therapy. The DCB used in this study had a paclitaxel-iopromid matrix coating.

For DCB angioplasty, The DCB extended beyond lesion margins by 2–3 mm and its diameter was 1:1 to the RVD. It was inflated under pressure of 8–10 atm with a total inflation duration up to 30–60 s. Each DCB catheter was used only once. Procedural success was defined as a residual stenosis less than 30 % and TIMI flow grade 3 without flow limiting dissection. In case of apparent dissection (type C or above) or TIMI flow less than grade 3, the bailout stenting with DES was performed.

2.6. Follow up and outcomes

After DCB angioplasty, clinical and angiographic follow up of all patients was lasted up to one year. The clinical follow up was performed by either phone interview or outpatient visit over the course of one year. The primary endpoint in this study was in-lesion late lumen loss (LLL). The secondary endpoints included composite MACE, cardiac death, non-fatal myocardial infarction (MI), TLR, target vessel revascularization (TVR), and vessel thrombosis. Cardiac death was defined if the cause was unknown or undeterminable. The MI was diagnosed according to recent guidelines by typical ischaemic symptoms, significant ECG changes, and significantly elevated troponin.¹² The ARC criteria were used to define vessel thrombosis.¹³

2.7. Statistical analysis

The Kolgormonov–Smirnov test was first used to assess the data distribution. The categorical variables were expressed as number (%). The chi-square test or Fisher exact test was used to assess significant differences between these categorical variables. The continuous variables were expressed as mean ± standard deviation (SD). The unpaired Student's t-test or Mann–Whitney U-test was used to assess significant differences between these continuous variables. The Kaplan–Meier curve for comparing composite MACE between SvCAD and LvCAD groups during one year follow up was drawn to estimate the statistical significance. A two-sided test was applied and a p value < 0.05 indicated a statistically significant difference. Data were analysed using SPSS 22.0 statistical software (IBM, Munich, Germany).

3. Results

3.1. Baseline characteristics

The baseline demographic characteristics were summarized in Table 1. Of these 237 patients, 133 (56.1 %) were males with the mean age of 59.73 ± 9.81 years. In this DCB-only study, 123 patients were included in SvCAD group and the other 114 patients were included in LvCAD group. There were no significant differences between both groups regarding the prevalence of cardiovascular risk factors. The incidence of ACS was comparable in both SvCAD (59.3 %) and LvCAD (61.4 %) groups with no significant difference noted (p = 0.75). There were no significant differences between both groups regarding other clinical and laboratory characteristics.

Table 1.

Baseline characteristics of the studied groups.

	All patients (n = 237)	SvCAD (n = 123)	LvCAD (n = 114)	p-value
Demographic characteristics
Male sex, n (%)	133 (56.1 %)	68 (55.3 %)	65 (57 %)	0.79
Age, years	59.73 ± 9.81	59.68 ± 10.15	59.77 ± 9.48	0.94
Cardiovascular risk factors, n (%)
Diabetes Mellitus	135 (57 %)	68 (55.3 %)	67 (58.8 %)	0.59
Hypertension	157 (66.2 %)	84 (68.3 %)	73 (64 %)	0.49
Family history of CADs	104 (43.9 %)	49 (39.8 %)	55 (48.2 %)	0.19
Current smoking	144 (60.8 %)	73 (59.3 %)	71 (62.3 %)	0.64
Dyslipidemia	123 (51.9 %)	62 (50.4 %)	61 (53.5 %)	0.63
CKD	69 (29.1 %)	33 (26.8 %)	36 (31.6 %)	0.42
Clinical characteristics
Clinical presentation, n (%)
ACS	143 (60.3 %)	73 (59.3 %)	70 (61.4 %)	0.75
CCS	94 (39.7 %)	50 (40.7 %)	44 (38.6 %)
Blood pressure, mmHg
SBP	142.57 ± 19.27	140.08 ± 19.8	145.26 ± 18.38	0.04
DBP	83.55 ± 11.8	83.11 ± 12.12	84.04 ± 11.48	0.54
Heart rate, b/m	78.48 ± 11.81	78.35 ± 12.15	78.61 ± 11.48	0.86
TTE parameters
LVEDD, mm	58.1 ± 7.65	57.89 ± 7.72	58.11 ± 7.61	0.82
LVESD, mm	36.82 ± 6.53	37.01 ± 6.55	36.62 ± 6.52	0.65
LVEF, %	51.9 ± 11.18	51.69 ± 11.46	52.13 ± 10.91	0.76
KILLIP class, n (%)
class I	186 (78.5 %)	101 (82.1 %)	85 (74.6 %)	0.26
class II	42 (17.7 %)	17 (13.8 %)	25 (21.9 %)
class III	9 (3.8 %)	5 (4.1 %)	4 (3.5 %)
Laboratory characteristics
Hemoglobin, gm/dl	11.29 ± 1.81	11.41 ± 1.87	11.16 ± 1.72	0.27
Leukocytes, x103/μL	8.22 ± 3.31	8.25 ± 3.39	8.18 ± 3.23	0.87
Platelets, x103/μL	297.52 ± 98.02	304.53 ± 96.67	289.96 ± 99.33	0.25
TGs, mmol/L	2.48 ± 1.21	2.35 ± 1.07	2.62 ± 1.33	0.08
TC, mmol/L	5.89 ± 1.19	5.9 ± 1.19	5.88 ± 1.2	0.92
LDL-C, mmol/L	3.25 ± 0.93	3.2 ± 0.98	3.3 ± 0.89	0.43
HDL-C, mmol/L	1.1 ± 0.33	1.09 ± 0.35	1.11 ± 0.31	0.67
Creatinine, mmol/L	94.41 ± 32.66	92.21 ± 31.77	96.76 ± 33.57	0.29
eGFR, ml/min/1.73 m2	53.49 ± 12.19	53.98 ± 13.43	52.96 ± 10.73	0.52

ACS: acute coronary syndrome; CADs: coronary artery diseases; CCS: chronic coronary syndrome; CKD: chronic kidney disease; DBP: diastolic blood pressure; DCB: drug coated balloon; DES: drug eluting stent; eGFR: estimated glomerular filtration rate; HDL-C: high density lipoprotein cholesterol; LDL-C: low density lipoprotein cholesterol; LvCAD: large vessel coronary artery disease; LVEDD: left ventricular end diastolic dimension; LVEF: Left ventricular ejection fraction; LVESD: left ventricular end systolic dimension; SBP: systolic blood pressure; SvCAD: small vessel coronary artery disease; TC: total cholesterol; TGs: triglycerides; TTE: trans-thoracic echocardiogram.

3.2. Lesion characteristics

The lesion characteristics were summarized in Table 2. Multi-vessel disease was similarly noted in both groups (27.6 % in SvCAD vs. 34.2 % in LvCAD, p = 0.27). The target lesions in the SvCAD group were more frequently located in the left anterior descending artery (LAD) (39.8 %) and the right coronary artery (RCA) (31.7 %). However, the left circumflex artery (LCX) was the main target vessel in the other LvCAD group (50.9 %) (p = 0.002). The RVD averaged at 2.89 ± 0.44 mm, and it was significantly lower in SvCAD group compared to that in LvCAD group (2.52 ± 0.29 vs. 3.25 ± 0.12 mm, p = 0.001).

Table 2.

Lesion and procedural characteristics of the studied groups.

	All patients (n = 237)	SvCAD (n = 123)	LvCAD (n = 114)	p-value
Lesion characteristics
Multi-vessel disease,n (%)	73 (30.8 %)	34 (27.6 %)	39 (34.2 %)	0.27
Target vessel, n (%)				0.002
LAD	83 (35 %)	49 (39.8 %)	34 (29.8 %)
LCX	93 (39.2 %)	35 (28.5 %)	58 (50.9 %)
RCA	61 (25.7 %)	39 (31.7 %)	22 (19.3 %)
Lesion length, mm	19.34 ± 6.15	19.14 ± 6.74	19.54 ± 6.29	0.93
RVD, mm	2.89 ± 0.44	2.52 ± 0.29	3.25 ± 0.12	0.001
Pre-PCI MLD, mm	1.05 ± 0.43	0.79 ± 0.33	1.31 ± 0.36	0.04
Pre-PCI diameter stenosis, %	85.25 ± 9.39	85.42 ± 9.85	85.0 ± 10.07	0.96
Procedural characteristics
Post-PCI MLD, mm	2.71 ± 0.44	2.33 ± 0.21	3.06 ± 0.25	0.001
Post-PCI diameter stenosis, %	12.68 ± 12.2	12.89 ± 13.69	12.46 ± 10.42	0.79
Acute gain, mm	1.71 ± 0.34	1.5 ± 0.26	1.92 ± 0.29	0.04
Follow up MLD, mm	2.77 ± 0.44	2.41 ± 0.21	3.13 ± 0.25	0.001
Procedure time, minutes	24.41 ± 6.32	23.67 ± 6.52	25.21 ± 6.02	0.06
Fluoroscopy time, minutes	16.12 ± 5.75	16.71 ± 5.72	15.49 ± 5.74	0.1
Amount of contrast, ml.	156.41 ± 46.01	159.92 ± 45.79	152.63 ± 46.14	0.22
Device length, mm	20.21 ± 6.07	20.02 ± 6.42	20.4 ± 6.44	0.93
Device diameter, mm	2.93 ± 0.47	2.56 ± 0.37	3.3 ± 0.16	0.003
Inflation pressure, atm	9.48 ± 2.21	9.38 ± 2.29	9.58 ± 2.37	0.89
Inflation duration, seconds	52.82 ± 13.73	51.82 ± 14.05	53.82 ± 14.98	0.83
Bail-out stenting, n (%)	7 (3 %)	3 (2.4 %)	4 (3.5 %)	0.63

DCB: drug coated balloon; DES: drug eluting stent; LAD: left anterior descending coronary artery; LCX: left circumflex coronary artery; LvCAD: large vessel coronary artery disease; MLD: minimal lumen diameter; PCI: percutaneous coronary intervention; RCA: right coronary artery; RVD: reference vessel diameter; SvCAD: small vessel coronary artery disease.

Regarding specific lesion characteristics, the lesion length and percentage of diameter stenosis were not statistically different between SvCAD and LvCAD groups (19.14 ± 6.74 vs. 19.54 ± 6.29 mm, p = 0.93 & 85.42 ± 9.85 % vs. 85.0 ± 10.07 %, p = 0.96, respectively). However, the MLD in SvCAD group was significantly lower than that in LvCAD group (0.79 ± 0.33 vs. 1.31 ± 0.36 mm, p = 0.04).

3.3. Procedural characteristics

The procedural characteristics were summarized in Table 2. Regarding DCB features, the mean diameter of DCB used in SvCAD group was significantly lower than that used in LvCAD group (2.56 ± 0.37 vs. 3.3 ± 0.16 mm, p = 0.003) with no significant difference concerning its length (20.02 ± 6.42 vs. 20.4 ± 6.44 mm, p = 0.93). The average inflation pressure (9.48 ± 2.21 atm) and duration (52.82 ± 13.73 s) were nearly comparable in both groups with no statistically significant difference (9.38 ± 2.29 vs. 9.58 ± 2.37 atm, p = 0.89 & 51.82 ± 14.05 vs. 53.82 ± 14.98 s, p = 0.83, respectively). Moreover, bail-out stenting was reported in 7 patients in this study (3 %); 3 patients in SvCAD (2.4 %) and 4 patients in LvCAD (3.5 %), with no significant difference noted between both group (p = 0.63).

Accordingly, after DCB intervention, the percentage of diameter stenosis was reduced to 12.68 ± 12.2 % with no significant difference noted between both groups (12.89 ± 13.69 % in SvCAD vs.12.46 ± 10.42 % in LvCAD, p = 0.79). On the other hand, the average MLD increased to 2.71 ± 0.44 mm and it was significantly higher in LvCAD group than in SvCAD group (3.06 ± 0.25 vs. 2.33 ± 0.21 mm, p = 0.001). Thus, the acute gain was 1.71 ± 0.34 mm and it was obviously greater in LvCAD group compared to SvCAD group (1.92 ± 0.29 vs. 1.5 ± 0.26 mm, p = 0.04). Furthermore, the MLD assessed during follow up was slightly greater than that assessed immediately after angioplasty and it was 2.77 ± 0.44 mm with greater value in LvCAD group than that noted in SvCAD group (3.13 ± 0.25 vs. 2.41 ± 0.21 mm, p = 0.001).

Regarding other procedural characteristics, the procedure time (24.41 ± 6.32 min), fluoroscopy time (16.12 ± 5.75 min), and amount of contrast (156.41 ± 46.01 ml) were not statistically different between both groups (23.67 ± 6.52 vs. 25.21 ± 6.02 min, p = 0.06 & 16.71 ± 5.72 vs. 15.49 ± 5.74 min, p = 0.1 & 159.92 ± 45.79 vs. 152.63 ± 46.14 ml, p = 0.22, respectively). The overall procedural success was achieved in most patients in both DCB arms.

3.4. Clinical outcomes

Clinical events during one year follow up were shown in Table 3. All 237 patients were available for monitoring long-term outcomes. The primary endpoint was in-lesion LLL; defined as follow-up minus post-PCI MLD, and it was negative in both DCB groups without significant difference (-0.07 ± 0.02 mm in SvCAD vs. - 0.06 ± 0.04 mm in LvCAD, p = 0.69) (Fig. 1).

Table 3.

Outcomes at one year follow up of the studied groups.

	All patients (n = 237)	SvCAD (n = 123)	LvCAD (n = 114)	p-value
Primary endpoints
In-lesion late lumen loss, mm	−0.07 ± 0.03	−0.07 ± 0.02	−0.06 ± 0.04	0.69
Secondary endpoints
Composite MACE, n (%)	20 (8.4 %)	8 (6.5 %)	12 (10.5 %)	0.27
Cardiac death, n (%)	2 (0.8 %)	1 (0.8 %)	1 (0.9 %)	0.96
Non-fatal MI, n (%)	14 (5.9 %)	6 (4.9 %)	8 (7 %)	0.49
TLR, n (%)	12 (5.1 %)	5 (4.1 %)	7 (6.1 %)	0.47
TVR, n (%)	7 (3 %)	3 (2.4 %)	4 (3.5 %)	0.63
Vessel thrombosis, n (%)	5 (2.1 %)	2 (1.6 %)	3 (2.6 %)	0.59

DCB: drug coated balloon; DES: drug eluting stent; LvCAD: large vessel coronary artery disease; MACE: major adverse cardiac event(s); MI: myocardial infarction; SvCAD: small vessel coronary artery disease; TLR: target lesion revascularization; TVR: target vessel revascularization.

During one year follow up, the secondary endpoints were observed including composite MACE, cardiac death, non-fatal MI, TLR, TVR, and vessel thrombosis. After DCB angioplasty, it was noted that 8.4 % of enrolled patients developed adverse clinical events in the form of composite MACE with no significant difference noted between SvCAD and LvCAD groups (6.5 % vs. 10.5 %, p = 0.27). Moreover, Kaplan–Meier analysis for composite MACE did not indicate a significant difference between SvCAD patients treated with DCB-only strategy and those with LvCAD (p = 0.276) (Fig. 3).

Fig. 3.

Kaplan–Meier curve of composite MACE at one year follow up.

There is no significant difference between SvCAD and LvCAD groups regarding composite MACE at one year follow up (P = 0.276).

MACE: major adverse cardiac event(s); LvCAD: large vessel coronary artery disease; SvCAD: small vessel coronary artery disease.

Specifically, this composite MACE was mainly related to non-fatal MI (5.9 %) and TLR (5.1 %) with a lower prevalence of cardiac death (0.8 %), vessel thrombosis (2.1 %), and TVR (3 %). In comparisons between SvCAD and LvCAD groups, the incidence of these adverse clinical events was not statistically significant accounting for 0.8 % vs. 0.9 % for cardiac death (p = 0.96), 4.9 % vs.7 % for non-fatal MI (p = 0.49), 4.1 % vs. 6.1 % for TLR (p = 0.47), 2.4 % vs. 3.5 % for TVR (p = 0.63), and 1.6 % vs. 2.6 % for vessel thrombosis (p = 0.59) (Fig. 2).

Fig. 2.

Outcomes DCB in SvCAD and LvCAD groups at one year follow up.

(a) Primary endpoints including late lumen loss. (b) Secondary endpoints including composite MACE, cardiac death, non-fatal MI, TVR, TLR, and vessel thrombosis.

(a) Primary endpoints: late lumen loss is not significantly different between SVD and LVD groups (P=0.69). (b) Secondary endpoints: composite MACE (P=0.27), cardiac death (P=0.96), non-fatal MI (P=0.49), TLR (P=0.47), TVR (P=0.63), and vessel thrombosis (P=0.59) are not significantly different between SvCAD and LvCAD groups

DCB: drug coated balloon; MACE: major adverse cardiac event(s); LvCAD: large vessel coronary artery disease; MI: myocardial infarction; SvCAD: small vessel coronary artery disease; TLR: target lesion revascularization; TVR: target vessel revascularization.

4. Discussion

We enrolled 237 patients in this clinical study to assess the safety and efficacy of DCB-only strategy in treating de novo lesions of coronary vessels, involving both small and large vessels. To our knowledge, previous studies focused mainly on the role of DCB in treating SvCAD with only few exclusive studies aiming at using DCB-only strategy in treating LvCAD. Accordingly, in this study, the percentage of SvCAD was 51.9 % (123 of 237 patients) and that of LvCAD was 48.1 % (114 of 237 patients).

The main findings of this study are as follows: (1) DCB-alone angioplasty is efficient and safe in treating not only SvCAD, but also LvCAD; (2) Late luminal enlargement of the culprit lesion during angiographic follow up was comparable in both SvCAD and LvCAD groups where in-lesion LLL was negative in both arms with no statistical significant difference; (3) The risk of composite MACE and other adverse clinical events in patients with LvCAD was not significantly different to those with SvCAD after DCB treatment. Thus, the use of DCB-only strategy in treating large coronary arteries leads to similar results and outcomes as compared to small arteries.

The DCB is a novel interventional modality based on rapid and homogenous transfer of an anti-proliferative drug from the balloon into vascular wall without implanting permanent metallic device. Recently, its safety and efficacy has been widely documented in treating CAD, especially SvCAD. According to recent DCB consensus, DCB-only strategy is increasingly regarded as a promising approach for treating de novo coronary lesions¹⁴ and could be further recommended as an alternative to DES in specific clinical situations as diabetes mellitus,¹⁵ and high bleeding risk (HBR).¹⁶

The first coronary lesions targeted mainly by DCB angioplasty were ISR and SvCAD. A multicenter randomized trials conducted by Jensen et al¹⁷ and Unverdorben et al¹⁸ reported that the ratio of TLR and LLL resulting from DCB and DES intervention used in treating ISR was not significantly different. Moreover, the coronary intervention guidelines recommended DCB intervention in treating ISR of both bare metal stent (BMS) and DES with a class IA recommendation.⁶ In addition, the efficacy of DCB in treating SvCAD has been supported by several studies as BASKET-SMALL 2,¹⁹ PICCOLETO²⁰ and the RESTORE SVD²¹ trials demonstrating the great therapeutic value of DCB in treating lesions in small coronary vessels. Furthermore, PICCOLETO II study has also reported the higher risk of three years MACEs in SvCAD patients treated with DES compared to those treated with DCB. These findings support a potential superiority of DCB-only approach in treating small coronary vessels.²²

However, data supporting the effectiveness of DCB in treating large coronary vessels are scarce. This is because large coronary arteries are more susceptible to recoil and dissection than small arteries due to more smooth muscle fibres leading to subsequent acute occlusion or restenosis, especially in the era of balloon angioplasty. Therefore, many interventional cardiologists had doubts about the safety of DCB-alone for large vessel lesions.²³ For this reason, doubts exist regarding the safety and feasibility of DCB-only strategy in treating large coronary vessels.

Accordingly, the main aim of this study was to effectively address the therapeutic potential of DCB-only strategy in treating LvCAD. Of 237 patients enrolled in this study and treated with DCB angioplasty, 114 patients had LvCAD and treated with DCB-only strategy. It was noted that MLD immediately after treating patients with small and large coronary vessels with DCB increased successfully with a significantly higher value regarding LvCAD (3.06 ± 0.25 vs. 2.33 ± 0.21 mm, p = 0.001). Besides, the percentage of diameter stenosis was reduced significantly after treating both small and large vessels with DCB with no significant difference noted between two arms (12.89 ± 13.69 % vs. 12.46 ± 10.42 %, p = 0.79). The resulting acute gain was obviously reported with a higher value in LvCAD group compared to SvCAD group (1.92 ± 0.29 vs. 1.5 ± 0.26 mm, p = 0.04).

Interestingly, compared with immediate MLD after DCB angioplasty (2.71 ± 0.44 mm), follow up coronary angiography done in this study revealed that this MLD (2.77 ± 0.44 mm) was further increased in all treated patients, mainly in those with LvCAD with a significant difference in comparison to SvCAD patients (3.13 ± 0.25 vs. 2.41 ± 0.21 mm, p = 0.001). This angiographic documentation of late positive remodelling or late luminal catch-up after DCB intervention was also consistent with the findings supported by Kleber et al²⁴

Despite these procedural outcomes favouring DCB angioplasty in treating lesions of large coronary vessels, the procedure and fluoroscopy times in addition to amount of contrast were not significantly different compared to DCB intervention of SvCAD (25.21 ± 6.02 vs. 23.67 ± 6.52 min, p = 0.06 & 15.49 ± 5.74 vs. 16.71 ± 5.72 min, p = 0.1 & 152.63 ± 46.14 vs. 159.92 ± 45.79 ml, p = 0.22, respectively). Moreover, this study was in line with the study stated by Yu et al where the incidence of coronary dissection after DCB intervention requiring bail-out stenting (0.07 ± 0.25 per DCB) was comparable in both SvCAD and LvCAD groups with no statistical difference (0.06 ± 0.23 vs. 0.08 ± 0.27 per DCB, p = 0.5).¹⁰

In the present study, the DCB-alone approach in LvCAD angioplasty showed similar angiographic outcomes when compared to SvCAD intervention. In terms of in-lesion LLL, there were not significant differences between SvCAD and LvCAD groups (-0.07 ± 0.02 vs. - 0.06 ± 0.04 mm, p = 0.69). This finding is in line with Valentines II; a study with a similar design to ours, where DIOR® paclitaxel DCB was used to treat 109 de novo coronary lesions.²⁵ The RVD was 2.40 ± 0.51 mm, but the percentage of included large vessels was not reported. Follow-up coronary angiogram at 6 months revealed that the in-lesion LLL (0.38 ± 0.39 mm) was significantly higher than that in the present study (-0.07 ± 0.03 mm), with no significant differences between SvCAD and LvCAD groups (p = 0.69). Moreover, Compared with the LLL achieved in BELLO study using IN.PACT Falcon DCB (0.08 ± 0.38 mm)²⁶ and PEPCAD I study using SeQuent Please® DCB (0.18 ± 0.38 mm)²⁷ where DCB used alone to treat SvCAD, the in-lesion LLL during follow up of SvCAD group included in the present study was also significantly lower (-0.07 ± 0.02), with no significant difference noted related to LvCAD group (-0.06 ± 0.04) (p = 0.69). Whether this difference in LLL was attributed to different DCB designs or a relatively short follow up period remains unclear.

In addition to these angiographic outcomes documented in the present study, the DCB-alone approach in LvCAD angioplasty showed similar long term safety regarding clinical outcomes when compared to SvCAD intervention. In Valentines II study, the incidence of MACE was 20.4 % during a follow-up period of 7.5 months.²⁵ However, in this study, the incidence of composite MACE was obviously lower (8.4 %) without significant difference between SvCAD and LvCAD arms (6.5 % vs. 10.5 %, p = 0.27). Moreover, the Kaplan–Meier curve of composite MACE at one year follow up revealed no significant difference between SvCAD and LvCAD groups (p = 0.276).

The overall incidence TLR in the present study was 5.1 % with no statistically significant difference noted between SvCAD and LvCAD groups (4.1 % vs. 6.1 %, p = 0.47). This relatively lower TLR rate obtained in this study, especially in LvCAD, could be attributed to good lesion preparation, proper DCB handling, and limited additional stent deployment. Concurrently, the overall TVR ratio reported in this study was 3 %, with no significant difference reported between SvCAD and LvCAD groups (2.4 % vs. 3.5 %, p = 0.63)

Previous studies using DCB-only strategy to treat coronary vessels revealed a relatively low risk of vessel thrombosis, even with only one moth DAPT.²⁸ In line with these findings, DAPT was prescribed in the present study after DCB angioplasty for at least 3 months and reported vessel thrombosis in 2.1 % of enrolled patients during one year follow up, with no significant difference noted between SvCAD and LvCAD arms (1.6 % vs. 2.6 %, p = 0.59). This vessel thrombosis could be linked to prolonged time needed for DCB to reach coronary lesions after entering the guiding catheter with subsequent more drug elution from the balloon and reducing the therapeutic DCB effect.²⁹ Thus, this DCB-only strategy confers a great clinical value regarding safety without the need for prolonged DAPT duration.

In addition, one year non-fatal MI rate was 5.9 %; 4.9 % in SvCAD and 7 % in LvCAD groups with no significant difference noted (p = 0.49). Most of these MIs could be related to enrolment of 60.3 % ACS patients in this study, and not necessarily linked to DCB treated coronary vessels.

On the other hand, the cardiac death rate reported during one year follow up was only 0.8 %; 0.8 % in SvCAD and 0.9 % in LvCAD groups with no significant difference (p = 0.96). Specifically, the rate of cardiac death in patients with LvCAD (0.9 %) is relatively lower than the rate of non-fatal MI (7 %) reported in this study. This cardiac death could be spontaneous and not attributed to DCB-only angioplasty as TLR and TVR ratios were 6.1 % and 3.5 % respectively in LvCAD patients.

In summary, the findings of the present study report very promising clinical implications regarding the use of DCB-only strategy in treating de novo lesions of large coronary arteries, with angiographic and clinical outcomes similar to those obtained in small coronary arteries.

5. Study limitations

This study had some limitations. First, it is a single-centre non-randomized study with a relatively small size of study population. Thus, multicentre prospective studies including larger study populations will be required to validate the clinical efficacy of DCB-only strategy in treating patients with large coronary vessels. Second, some ascertainment bias might not be totally excluded. However, data processing and analysis were performed by an independent statistician. Third, these results could not be reproducible in different centres as they were achieved in a qualified centre frequently using DCB angioplasty. Fourth, the ideal way to document the efficacy of DCB in LvCAD is through a direct head-to-head comparison of DCB and DES in this setting. Thus, future studies that directly compare DCB and DES in LvCAD will be needed to strengthen the efficacy of using DCB in these clinical settings. Finally, the follow up duration is relatively short and a longer one could be considered to effectively investigate long term angiographic and clinical outcomes.

6. Conclusion

This study documented the safety and efficacy of DCB-only strategy in treating de novo large coronary vessel with comparable clinical events and angiographic outcomes to small coronary vessels during one year follow-up. These findings underline the concept that coronary angioplasty of de novo lesions without implanting permanent metallic devices is less dependent on vessel diameter and could be used as a first choice therapeutic modality for patients not eligible for stenting.

Sources of funding

None.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.