FormalPara Key Summary Points

We did not find a significant difference in the rate of slow flow/no-reflow between the traditional rota-flush group and heparin rota-flush group (19.6% vs. 29.1%, P = 0.117), and the procedural success was also comparable.

The incidence of coronary spasm was significantly higher in the heparin rota-flush group compared to the traditional rota-flush group (42.7% vs. 22.7%, P = 0.003).

Although the incidence of systolic blood pressure (SBP) drop ≥ 20 mmHg was significantly higher in the traditional rota-flush group, severe hypotension (SBP < 90 mmHg) was not significantly different (P = 0.841), indicating that vasodilators in the flush did not severely impact hemodynamic stability.

Introduction

Rotational atherectomy (RA) was first utilized to treat obstructive coronary diseases by Jarome Ritchie and colleagues in 1987 [1], and thereafter has been widely applied to modify plaque compliance and allow smooth passage for stents in highly calcific lesions of the left main coronary artery, bifurcations, ostial side branches, and chronic total occlusions [2,3,4,5,6]. Despite the high efficacy for lesions not suitable for direct coronary angioplasty or stenting, severe complications including coronary perforation (0–1.5%), dissection (10–13%), severe spasm (0.6–1.6%), and slow flow/no-reflow (1.2–7.6%) may occur during the RA procedure [7]. Distal coronary spasm can lead to adverse events such as ventricular fibrillation, and complete atrioventricular block [8]. Moreover, RA has been associated with a higher incidence of slow flow/no-flow than other coronary intervention measures, a phenomenon, if not timely handled, that can result in serious ischemic complications such as myocardial infarction, cardiogenic shock, or even death [9]. Many factors influence the incidence of slow flow/no-reflow related to the RA procedure, including lesion length, lesion angulation, reference diameter, total ablation time, short single run (< 15 s), systolic blood pressure (SBP) > 140 mmHg, and initial burr-to-artery ratio [10]. Traditionally, a “cocktail” consisting of normal saline, unfractionated heparin, verapamil, nitroglycerin, and RotaGlide lubricant was utilized to prevent slow flow/no-reflow during RA. However, the use of RotaGlide is contraindicated in patients with known allergies to it and adds to the cost of the RA procedure. A retrospective study by Whiteside et al. reported that similar outcomes were achieved despite not using RotaGlide as a part of the rota-flush solution [11]. Additionally, vasodilators added to the rota-flush solution can lead to hemodynamic disturbances like hypotension, bradycardia, and transient atrioventricular block [12]. There is growing evidence supporting simplifying traditional rota-flush as a heparin saline rota-flush. Siddiqi et al. reported that combining modern procedural techniques with the routine use of heparin-only rota-flush solution, RA procedures were safe with low slow flow/no-reflow rates (2.2%) and high rates of procedural success (94.8%) [13]. The aforementioned study was limited by its retrospective nature and small sample size. We conducted this prospective randomized controlled trial to compare the feasibility and safety of the heparin rota-flush and traditional rota-flush in patients undergoing RA.

Methods

Study Design and Patient Selection

This single-center, randomized, triple-blind, parallel-group trial was conducted at the Department of Cardiology, the Affiliated Anhui Provincial Hospital of Anhui Medical University from August 2019 to November 2021. Patients older than 18 years, with an indication for RA were selected consecutively in the catheterization laboratory. The inclusion criteria were as follows: (1) angiographically moderate or severe calcific lesions; (2) diffuse lesions expected to be difficult to stent; (3) ostial or bifurcation lesions; (4): intravascular ultrasound (IVUS) indicating a large arc of superficial calcium involving three or four quadrants; (5) RA as a secondary approach after unsuccessful attempt to dilate a lesion or failure to deliver stents. The exclusion criteria were (1) bradycardia (heart rate ≤ 60 bpm) or peripheral blood pressure ≤ 90/60 mmHg; (2) atrioventricular block; (3) target lesions were culprit vessels of acute myocardial infarction within 4 weeks; (4) the burr cannot cross the target lesion; (5) history of coronary artery bypass grafting (CABG); (6) lesion angulation > 90°; (7) inability to provide informed consent. The study conformed to the principles of the Helsinki Declaration and the institutional review board of the Affiliated Anhui Provincial Hospital of Anhui Medical University approved the study protocol. All patients provided written informed consent before procedures.

Randomization

Randomization was performed using a computer-generated list of random numbers; the allocation sequence was concealed from patients, the RA operators, and angiogram analyzers in sequentially numbered, opaque, sealed, stapled envelopes. Patients included in the study were randomly allocated (1:1) into the heparin rota-flush and the traditional rota-flush groups. Patients in the former group received a rota-flush solution containing 10,000 U unfractionated heparin in 1 L saline, while patients in the latter group received a cocktail containing 10,000 U unfractionated heparin, 10 mg verapamil, and 4 mg nitroglycerin in 1 L saline.

Revascularization Protocol

A radical, brachial, or femoral approach was employed on the basis of peripheral vascular condition and procedural requirement. In the catheterization laboratory, all patients in this study were preloaded with dual antiplatelet therapy. RA was performed with the Rotablator system (Boston Scientific Corporation, Natick, MA, USA). The pressured rota-flush solution was continuously infused into the coronary artery through a 4Fr Teflon sheath of the Rotablator system. Initial RA burr size was either 1.25 or 1.5 mm according to the senior operator’s selection; the burr was subsequently advanced to a position proximal to the lesion, the initial rotational speed was set within the conventional range (140,000–180,000 rpm), and the burr was activated and moved forward with a slow pecking motion. Each run time was less than 30 s, and care was taken to avoid a decrease in rotational speed of more than 5000 rpm. Following RA, routine balloon dilatation to facilitate drug-eluting stent (DES) implantation or drug-coated balloon (DCB) employment was performed unless serious complications like coronary perforations happened. If the slow flow/no-reflow phenomenon occurred, an intracoronary infusion of verapamil and/or sodium nitroprusside was immediately administered to improve coronary flow, and an intravenous infusion of vasoactive drugs such as noradrenalin or atropine was administered to maintain hemodynamic stability if necessary. Furthermore, an intra-aortic balloon pump (IABP) was used in high-risk cases with severe left ventricular dysfunction, unprotected left main stenosis, and severe three-vessel diseases. IVUS use was left to the operator’s discretion.

Definitions and Endpoints

The primary endpoint of the study was the incidence of slow flow/no-reflow after RA. The secondary endpoints of the study were procedural success, RA-related procedural complications, and in-hospital major adverse cardiovascular events (MACE). Slow flow was defined as the diminution of coronary flow by 1–2 thrombolysis in myocardial infarction (TIMI) flow grades from the baseline antegrade flow, and no-reflow was defined as the cessation of blood flow into the distal coronary artery in the absence of dissection or thrombus immediately after RA. Procedural success was defined as the achievement of a residual diameter stenosis less than 30%, TIMI flow grade 3, and in the absence of cardiac death, coronary perforation, stroke, stent-thrombosis (ST), and target-lesion revascularization (TLR). Vessel perforation (defined as any contrast extravasation) from the burr, burr entrapment, transection of the RA guidewire, and coronary dissection were included in RA-related complications. Coronary spasm was characterized by transient total or subtotal occlusion or severe diffuse vasoconstriction of an epicardial artery. According to National, Heart, Lung, and Blood Institute (NHLBI) criteria, dissection was defined as grades C to F [14]. Hypotension was defined as a drop of SBP ≥ 20 mmHg or SBP < 90 mmHg; bradycardia was heart rate < 60 bpm during the RA procedure. MACE was defined as the composite of TLR, ST, stroke, and cardiac death. Death was to be considered cardiac in origin unless a non-cardiac origin was confirmed. Thrombus of the target lesion on either angiography or autopsy examination was defined as ST according to the Academic Research Consortium [15]. TLR was defined as unplanned ischemia-driven percutaneous or surgical revascularization of the treated vessel.

Sample Size

On the basis of a clinical audit in our hospital and a literature review, we estimated that the incidence of slow flow/no-reflow during RA would be 4.0% [16] in the traditional rota-flush group and 15.0% [17, 18] in the heparin rota-flush group. A minimum of 194 patients (97 in each group) would provide 84.5% power to detect a difference between the two groups in the proportion of subjects with slow flow/no-re-flow. The test statistic used was the one-sided Z test and the significance level of the test was 0.05.

Statistical Analysis

After randomization, we recorded baseline characteristics of enrolled participants including demographic data, cardiovascular risks, angiography, and interventional procedure characteristics. The TIMI grade of coronary flow was confirmed by injecting sufficient contrast medium immediately after the burr was removed. Detail lesion information was documented by two cardiologists facilitated by a quantitative coronary angiography analysis system using offline, computer-based QAngio XA 7.3 software (MEDIS Imaging Systems, Leiden, the Netherlands). Continuous and categorical variables were expressed as mean ± SD and frequencies, respectively. The Kolmogorov–Smirnov test was performed to determine if the continuous variables were normally distributed. Normally distributed continuous variables were compared using unpaired Student’s t test; otherwise, the Mann–Whitney U test was employed. The categorical variables were compared using chi-square test or Fischer’s exact test. Univariate and multivariate logistic regression analyses were used to evaluate the independent predictors of slow flow/no-reflow and coronary spasm. All variables with a P ≤ 0.1 in the univariate analysis were included to the multivariate analysis model. The Hosmer–Hemeshow goodness-of-fit statistics showed that the model adequately fitted the data. All tests were two-tailed and P < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS (version 26.0, IBM, Armonk, New York, USA).

Results

A total of 220 patients met the inclusion criteria, and cases of acute myocardial infarction (ST-segment elevation myocardial infarction, STEMI n = 10; non-STEMI, NSTEMI n = 8), history of CABG (n = 1), and the burr would not cross (n = 1) were excluded from the study. Thus, a total of 200 patients were randomly assigned to one of the two groups: 103 to the heparin rota-flush group and 97 to the traditional rota-flush group. A flowchart demonstrating patient recruitment for the study is shown in Fig. 1. There were no significant differences in baseline clinical and angiographic characteristics as shown in Table 1. Vital signs including heart rates and systolic and diastolic blood pressure just before the RA procedure were comparable in the two groups. No significant difference of procedural characteristics was observed in the two groups as shown in Table 2. Procedural success was comparable between the two groups (97.1% vs. 93.8%, P = 0.320). A comparison of complications is presented in Table 3. A relatively higher incidence of slow flow/no-reflow was observed in the heparin rota-flush group without reaching statistical difference (29.1% vs. 19.6%, P = 0.117). Fewer patients developed hypotension in the heparin rota-flush group compared to the traditional rota-flush group (54.4% vs. 69.1%, P = 0.033), and the incidence of bradycardia tended to be less (20.4% vs. 24.7%, P = 0.461). However, the incidence of coronary spasm was significantly higher in the heparin rota-flush group (42.7% vs. 22.7%, P = 0.003). Other procedure complications were not different between the two groups. None of the 200 patients had transection of the RA guidewire. MACE including cardiac death, TLR, and ST were not different between the two groups. None of the 200 patients developed stroke. The univariate and multivariate logistic regression analyses to investigate underlying factors of slow flow/no-reflow are shown in Table 4. Among 11 variables, SBP just before the RA procedure (OR 0.82, 95% CI 0.67–0.99, P = 0.040) and reference diameter (OR 0.12, 95% CI 0.02–0.59, P = 0.010) were inversely associated with slow flow/no-reflow, while lesion length (OR 1.05, 95% CI 1.01–1.09, P = 0.027) and lesion angulation greater than 45° (OR 2.15, 95% CI 1.01–4.62, P = 0.049) were significantly associated with slow flow/no-reflow. After adjustment for the other 10 variables, the use of heparin rota-flush solution was not an independent factor for slow flow/no-reflow (OR 1.66, 95% CI 0.78–3.52, P = 0.186). In addition, as shown in Table 5, utilizing of the heparin rota-flush solution (OR 2.68, 95% CI 1.43–5.05, P = 0.002) was significantly associated with coronary spasm after adjusting for lesion length, reference diameter, mean RA speed, mean single run time, and initial burr-to-artery ratio.

Fig. 1
figure 1

Study flowchart. A total of 220 patients undergoing RA at the Affiliated Anhui Provincial Hospital of Anhui Medical University between August 2019 and November 2021 were selected consecutively; 20 patients were excluded from this study. Finally, 200 patients were enrolled and prospectively randomized into the traditional rota-flush group (N = 97) and the heparin rota-flush group (N = 103). NSTEMI non-ST-segment elevation myocardial infarction, CABG coronary artery bypass grafting, RA rotational atherectomy

Full size image
Table 1 Patients and angiographic characteristics
Full size table
Table 2 Procedural characteristics
Full size table
Table 3 Comparison of procedural success, complications, and MACE
Full size table
Table 4 Univariate and multivariate logistic regression analyses for slow flow/no-reflow
Full size table
Table 5 Univariate and multivariate logistic regression analyses for coronary spasm
Full size table

Discussions

The principal findings of this analysis are as follows:

  1. 1.

    Continuous intracoronary infusion of heparin rota-flush solution does not increase the incidence of slow flow/no-reflow compared to the traditional rota-flush solution.

  2. 2.

    Traditional rota-flush solution prevents coronary spasm more effectively compared to the heparin rota-flush solution.

  3. 3.

    Although the incidence of SBP drop ≥ 20 mmHg was significantly higher in the traditional rota-flush group, severe hypotension (SBP < 90 mmHg) was not significantly different (P = 0.841), and the usage of vasopressors and atropine was also similar between the two groups, indicating that vasodilators in the flush did not severely impact hemodynamic stability.

Several retrospective studies investigating the efficacy of heparin rota-flush in the RA procedure obtained similar findings to our study. A retrospective study aiming to evaluate the feasibility and safety of heparin rota-flush solution was performed by Lee et al. [19] and included 67 consecutive patients who underwent RA. As a result, procedural success was achieved in all patients, slow flow/no-reflow occurred in 7%, with subsequent resolution after intracoronary vasodilator therapy, indicating that the use of a heparin rota-flush solution was a feasible and reasonable alternative to a rota-flush solution that contains RotaGlide and vasodilators. More recently, in a retrospective study of 48 patients who underwent RA, procedural success was achieved in all patients with similar slow flow/no-reflow rates (12% vs. 4.3%, P = 0.330) [17]. The aforementioned two studies were both non-randomized and the sample size was relatively small. To the best of our knowledge, the present study was the first prospective randomized trial to compare the feasibility and safety of heparin rota-flush and traditional rota-flush solution in preventing slow flow/no-reflow followed by RA.

The statistically non-significant but numerically higher rates of slow flow/no-reflow in the heparin rota-flush group were likely due to four factors, absence of vasodilators, lower SBP before RA, more diffuse lesions, and lesion angulation greater than 45°. But after the multivariate logistic regression analysis, we found that the use of the heparin rota-flush solution was not independently associated with slow flow/no-reflow (OR 1.66, 95% CI 0.78–3.52, P = 0.186), suggesting that the addition of vasodilators did not have a statistically meaningful impact on slow flow/no-reflow. Note that the reported incidence of slow flow/no-reflow in the traditional rota-flush group was similar to that reported by Sakakura et al. (19.6% vs. 12%), but the incidence was significantly higher in the heparin rota-flush group (29.1% vs. 4.3%). The baseline clinical and procedural characteristics were comparable between the two groups in our study. The multivariate logistic regression analysis revealed additional elements involving the slow flow/no-reflow phenomenon; as was shown in Table 4, lesion length and lesion angulation greater than 45° were significantly associated with slow flow/no-reflow, whereas SBP just before the RA procedure and reference diameter were inversely associated with slow flow/no-reflow, which was consistent with the findings of Sakakura et al. [10].

Lesion length was a known risk factor for slow flow during RA [20], probably because the massive distal embolization of atheromatous debris would be greater in the cases of diffuse long lesions than that with short lesions. Vasodilator drugs like nitroglycerin and verapamil had theoretical benefit of reducing microvascular obstruction as evidenced by relatively lower rates of slow flow/no-reflow in the traditional rota-flush group (19.6% vs. 29.1%, P = 0.117). Angulation  greater than 45° was also closely associated with slow flow/no-reflow, one possible explanation being that RA operators would have difficulty in advancing the burr beyond the angle [21], resulting in longer ablation time and subsequent slow flow. Systolic blood pressure was inversely associated with slow flow/no-reflow, since low systolic blood pressure might reflect poor cardiac function; it was important to keep systolic blood pressure ≥ 120 mmHg (at least 100 mmHg) as advised by clinical expert consensus [10]. Besides, since the most frequent burr size was 1.5 mm in the present study, small vessel size less than 2.5 mm would naturally have a high (0.6) burr-to-artery ratio, a known risk factor for slow flow [22], which accounted for the finding that reference vessel diameter was inversely associated with slow flow. All patients who experienced slow flow/no-reflow were reverted by intracoronary administration of nitroglycerin, verapamil, tirofiban, or nitroprusside, and angiography demonstrated TIMI 3 flow grade prior to departure from the catheterization laboratory.

Although the use of the heparin rota-flush solution did not result in statistically significant higher incidence of slow flow, coronary spasm was more frequently observed (42.7% vs. 22.7%, P = 0.003). We performed the logistic regression analysis to identify the risk factors for coronary spasm in the heparin rota-flush group, and the results revealed that the use of heparin saline was significantly associated with coronary spasm (OR 2.68, 95% CI 1.43–5.05, P = 0.002). Spasm of coronary arteries may occur spontaneously or following interventional provocation, e.g., RA was known to provoke coronary spasm occasionally, and intracoronary administration of nitroglycerin and calcium channel antagonists were feasible rationales for preventing or resolving coronary spasm [23, 24]. The spasmed coronaries were all resolved completely after intracoronary injection of nitroglycerin or verapamil in our study. Bradycardia and atrioventricular block were more prevalent in the intervention of the right coronary artery (RCA) and left circumflex (LCX) coronary artery. In our study, two patients in the setting of RCA intervention developed transient complete atrioventricular block and reverted to sinus rhythm after stopping the RA procedure. Sinus arrest was documented in two cases each in the setting of RCA and LCX intervention. Out of the 67 patients from the traditional rota-flush group who developed hypotension, 15 patients had associated episodes of bradycardia in the setting of RCA or LCX intervention. The other 52 developed hypotension in the setting of left anterior descending (LAD) intervention, which was likely attributed to vasodilators in rota-flush solution. Although the incidence of SBP drop ≥ 20 mmHg was significantly higher in the traditional rota-flush group, severe hypotension (SBP < 90 mmHg) was not significantly different, and the usage of vasopressors was similar between the two groups, indicating that vasodilators in the flush did not severely impact hemodynamic stability. RA procedure failed in nine cases. Among three failed cases in the heparin rota-flush group, two cases developed coronary perforation in the setting of severely calcified LAD intervention and another case received emergent CABG after severely LCX dissection. Six cases failed in the traditional rota-flush group, four of them developed coronary perforation in the LAD intervention, one case experienced burr entrapment in the fibrocalcific lesion of the middle LAD, another failed case experienced persistent angina following successful RA and stent implantation, emergent coronary angiogram (CAG) revealed 80% stenosis due to acute stent thrombosis and the patient underwent successful PCI. Two cases in the heparin rota-flush group developed coronary perforation due to high pressure balloon dilatation. Among four cases in the traditional rota-flush group who developed coronary perforation, two cases resulted from high-speed RA of severely calcified and eccentric lesion in the LAD, and another two cases resulted from poor management of the guidewire. There was no transection of RA guidewire in the present study. Burr entrapment occurred in one case in the traditional rota-flush group and was removed successfully. In total, there were two deaths in our study population; both of the patients developed coronary perforation and received bailout CABG and died before discharge.

Limitations

This study was conducted at a single center with relatively small sample size, and the duration of follow-up was too short. The outcomes of our study need validation in a larger population of patients from multicenter and longer-term follow-up. Besides, periprocedural myocardial infarction (PMI) have been underdiagnosed, as cardiac biomarkers were not routinely tested in all patients. PMI, defined as an elevation of creatine kinase (CK) levels at least two times the upper limit of normal (ULN) with an elevation of CK-myocardial band (MB) levels above the ULN, was evaluated only in 154 patients in the present study. Ten of 74 patients (13.5%) in the heparin rota-flush group and 13 of 71 patients (18.3%) in the traditional rota-flush group developed PMI (P = 0.429), and the CK-MB level after the procedure was relatively lower in the former group without reaching statistical significance (28.4 ± 21.5 IU/L vs. 31.7 ± 26.4 IU/L, P = 0.407). Although a few studies suggested the determinants of PMI in RA were complex lesion features, such as diffuse lesion, severe calcification, and larger lesion angle, whether the rota-flush solutions correlate with the PMI is still unknown and needs validation by more studies in the future.

Conclusions

The findings suggest that although continuous intracoronary infusion of heparin rota-flush solution does not increase the incidence of slow flow/no-reflow, traditional rota-flush solution without RotaGlide prevents coronary spasm more effectively compared to heparin rota-flush without significant impact on severe hypotension. These results do not support a strategy of routine use of heparin rota-flush solution in patients receiving RA procedures.