Between January 2015 and December 2021, 579 consecutive patients who received RA therapy for severely calcified coronary lesions were retrospectively screened in our institution. Inclusion criteria were as following: (1) The length of calcified lesions $>$30 mm; (2) Multivessel coronary artery disease (CAD) with $\ge$70% diameter stenosis; (3) LVEF 40%. Patients who were hemodynamically unstable, presenting with ST-segment elevation myocardial infarction (STEMI), or patients refused to receive RA were excluded. Finally, IABP was inserted in 80 of the 596 patients before the RA procedure, and they were included in the IABP group. Analysis of propensity score matching (PSM) was applied to reduce the potential effect of bias based on propensity score of each patient. After PSM, 160 patients undergoing RA (80 patients in each study group) were matched in the field of multivessel coronary disease, the length of calcified lesions $>$30 mm, and LVEF. 4 patients who received a bailout IABP implantation were excluded. Finally, 156 patients were included in the present study, of whom 80 were in the IABP group and 76 were in the non-IABP group, separately (Fig. 1). The Institutional Review Board approved the data collection procedure of the study and all participants signed informed consent before RA procedure.

Fig. 1.

Study flow chart. Abbreviation: RA, rotational atherectomy; LVEF, left ventricular ejection fraction; IABP, intra-aortic balloon pump.

All RA procedures were performed by three senior experienced interventional cardiologists with the Rotablator system (Boston Scientific Corporation, Natick, MA, USA). The arterial access site was chosen based on peripheral vascular conditions and procedural requirements. Initial RA burr size was either 1.25 mm, 1.5 mm, or rarely 1.75 mm according to senior operators’ selection, then the burr was advanced proximally to the lesion, and moved forward with a slow pecking motion. The initial burr speed was set within the range from 140,000 to 180,000 rpm with the duration of each run less than 30 s, and a decrease in rotational speed $>$5000 rpm was carefully avoided. To reduce the occurrence of slow flow/no reflow, a pressured Rota-flush solution consisting of heparin, verapamil, and nitroglycerin was continuously infused into the coronary artery through a 4Fr Teflon sheath of the Rotablator system. An independent experienced cardiologist assessed the presence of slow-flow/no re-flow phenomenon by injecting a sufficient contrast medium immediately after the ablation pass. Following RA, routine balloon predilation to facilitate Drug-eluting stents (DES) implantation was performed. The IABP was placed percutaneously via the femoral artery, and 1:1 electrocardiographic triggering was initiated before starting the RA. Before removal of IABP, the electrocardiographic triggering was gradually down regulated from 1:1 to 1:2 to 1:3. The time to remove IABP was mainly determined by the patient’s clinical status (usually 4 to 24 hours following PCI). The decision to insert an IABP was left to the discretion and guidance of the supervising cardiologists. All patients received pretreatment with 300 mg aspirin and a loading dose of P2Y12 inhibitor (clopidogrel or ticagrelor) prior to RA, as well as the secondary prevention of CAD after the procedure. Cardiac biomarkers (Troponin I) were measured before PCI, and 6, 12, and 24 h after the RA procedure.

Severely calcified lesions were either visually assessed by coronary angiography, defined as radiopacities noted without cardiac motion before contrast injection, or Intra-vascular ultrasound (IVUS) indicated superficial calcium involving more than 3 quadrants. Planned RA was defined as RA performed directly before balloon predilation, while bailout RA was RA performed after failure to balloon predilation or stents deliver to target lesions. Slow flow/no re-flow was defined as less than Thrombolysis in Myocardial Infarction (TIMI) III flow grade in the absence of dissection or thrombus immediately after RA. A final residual stenosis 30% complied with TIMI flow grade III after stents placement was considered procedural success. The procedure was considered a failure if patients received emergent coronary artery bypass grafting (CABG) and/or PCI, or other severe RA-related complications (death, coronary perforation) developed before discharge. Periprocedural myonecrosis was defined as troponin I above threefold of the upper limit of normal or a 50% increase from the baseline level [7].

All patients were closely followed at 90-day and 180-day intervals after discharge. Follow-up information was obtained by clinicians through outpatient clinic visits, phone interviews, and hospital medical records. The primary endpoints of the present study included procedural success, and in-hospital major adverse cardiovascular events (MACE). MACE consisted of cardiac death, heart failure, target vessel revascularization (TVR), and stent thrombosis (ST). Unless a non-cardiac origin was surely documented, death was considered to be cardiac in origin. Deterioration in signs and symptoms of in patients with previous chronic heart failure (CHF) or new-onset heart failure (HF) requiring urgent therapy was considered as in-hospital HF. Diagnostic criteria was based on an intravenous administration of diuretic drugs, vasodilators, or inotropic drugs, and including at least one of the followings: cardiac pulmonary edema or pulmonary vascular congestion on chest radiograph; rales $>$one-third of the lung fields due to HF; left ventricular end-diastolic pressure (LVEDP) $>$18 mmHg; or dyspnea, with a Po2 80 mmHg or an oxygen saturation 90% without oxygen inhaled (significant lung disease excepted). TVR was defined as any repeat PCI or CABG of the target vessel due to stent thrombosis or perforation. Thrombus of the target lesion on either angiography or autopsy examination was considered as ST according to the Academic Research Consortium [8]. The secondary endpoints consisted of all-cause mortality and readmission due to HF at 90- and 180-day intervals after discharge. Readmission due to HF was defined as readmission primarily for the treatment of HF needing the use of intravenous therapy such as diuretics, inotropic agents, or vasodilators.

The SPSS 26.0 system (IBM, Armonk, NY, USA) was utilized for statistical calculations. A logistic model was used to calculate the probability of receiving a IABP support before RA procedure (the propensity score). Baseline characteristics including age, male, hypertension, diabetes mellitus (DM), atrial fibrillation (AF), history of HF, pre-MI, pre-PCI, chronic kidney disease (CKD), LVEF, NT-proBNP, systolic blood pressure (SBP) before RA, target vessel (LAD, LCX, or RCA), and diseased vessels (two or three) were set as covariates. Based on the propensity score in a 1:1 (IABP:Non-IABP) fashion, the nearest neighbor matching was performed with a maximum caliper of 0.2. Categorical variables were reported as value (percentage) and Chi-squared or Fisher’ exact test was utilized. If the continuous variables were normally distributed determined by the Wilk-Shapiro test, they were reported as a mean $\pm$ SD, and intergroup differences were compared using an unpaired Student’s t test. Otherwise, non-normal distribution data was shown as median [25th–75th quartiles], and intergroup differences were compared using a Mann-Whitney U test. In addition, we compared the cumulative incidence of all-cause mortality, 90-day and 180-day readmission due to HF using the Kaplan-Meier method and the log-rank test. To identify the influential factors for 90-day and 180-day readmission due to HF, a Cox regression model was performed. All reported p values were 2 tailed, and intergroup differences were considered statistically significant when the probability was 0.05.

Baseline demographics, comorbidities, and results of laboratory test were presented in Table 1. More frequent history of prior MI (26.3% vs. 7.9%, p = 0.002), more often CHF (42.5% vs. 5.3%, p 0.001), higher level of low density lipoprotein-cholesterol (LDL-C) (2.1 $\pm$ 0.8 mmol/L vs. 1.8 $\pm$ 0.6 mmol/L, p = 0.036) and NT-proBNP [1024.0 (201.3–2684.3) pg/mL vs. 284.0 (75.8–904.5) pg/mL, p 0.001] were observed in the IABP group. Fewer patients in the IABP group received nitrates and calcium channel blockers than that in the non-IABP group. However, no significant difference was observed with regard to other comorbidities, laboratory test results, and medications. Vital signs including baseline pressure and heart rates (HR), were also comparable.

Table 2 showed angiographic and procedural characteristics. The incidence of procedural success was similar (94.7% vs. 96.3%, p = 0.714) in the two groups. Of note, higher post-procedural SBP was observed in the IABP group (112.7 $\pm$ 22.5 mmHg vs. 94.3 $\pm$ 14.8 mmHg, p 0.001), there was no significant difference in the incidence of vasopressors usage in the IABP and non-IABP group (8.8% vs. 11.8%, p = 0.525). Moreover, lesion and other procedural characteristics showed no significantly difference between the two groups.

Table 3 summarized outcomes of in-hospital, 90-day and 180-day follow-up. Clinical follow-up was accomplished in all cases. Hypotension was less frequently observed in the IABP group (13.8% vs. 53.9%, p 0.001), and there was a trend towards less frequent slow flow/no re-flow (35.0% vs. 46.1%, p = 0.160) in this group. Other periprocedural complications including bradycardia, complete atrioventricular block, dissection, perforation, and coronary spasm were not significantly different in the two groups. No patients developed sinus arrest and burr entrapment in this study. The admission days were significantly shorter in the IABP group than in the non-IABP group (5.6 $\pm$ 1.0 vs. 7.1 $\pm$ 2.9, p 0.001).

Compared to the non-IABP group, in-hospital MACE was less frequently observed in the IABP group (7.5% vs. 26.3%, p = 0.002), mainly driven by in-hospital HF (6.3% vs. 23.7%, p = 0.002), as shown in Table 3. Compared to the non-IABP group, the incidence of periprocedural myonecrosis tended to be lower (27.5% vs. 34.2%, p = 0.364). No significant difference as for cardiac death and TVR were observed between the two groups, and stent thrombosis was observed in neither group.

The Kaplan-Meier analysis showed a significantly lower incidence of readmission due to HF in the IABP group during the 90-day follow-up (log-rank test: p = 0.002, HR = 0.32, 95% CI: 0.17–0.61, Fig. 2A).

Fig. 2.

Kaplan-Meier curves estimate incidence of readmission due to HF for patients undergoing elective RA with and without IABP support. (A) Kaplan-Meier curves of cumulative incidence of readmission due to HF within 90-day follow-up. (B) Kaplan-Meier curves of cumulative incidence of readmission due to HF within 180-day follow-up. Abbreviations: IABP, intra-aortic balloon pump; HF, heart failure; RA, rotational atherectomy; HR, hazard ratio.

In addition, the incidence was also significantly lower (log-rank test: p = 0.013, HR = 0.48, 95% CI: 0.27–0.88, Fig. 2B) during the 180-day follow-up.

Furthermore, Kaplan-Meier analysis showed that the cumulative survival rates within 90-day follow up were not different between the two groups (p = 0.274, Fig. 3).

Fig. 3.

Kaplan-Meier curves for cumulative survival rates within 90-day follow-up. Abbreviations: HR, hazard ratio; IABP, intra-aortic balloon pump.

Fig. 4 summarized the incidence of in-hospital HF, readmission due to HF at 90-day and 180-day intervals.

Fig. 4.

Incidence of in-hospital HF, readmission due to HF at 90-day and 180-day intervals. Abbreviations: IABP, intra-aortic balloon pump; HF, heart failure.

Cox multivariate analysis was performed to investigate influential factors of readmission due to HF during the 90-day and 180-day follow-up. The analysis determined that IABP support (HR = 0.34, 95% CI: 0.15–0.76, p = 0.008), in-hospital HF (HR = 3.28, 95% CI: 1.29–8.36, p = 0.013), and periprocedural myonecrosis (HR = 4.26, 95% CI: 1.60–11.35, p = 0.004) were independently associated with readmission due to HF within 90-day follow up (Table 4).

Additionally, IABP implantation (HR = 0.47, 95% CI: 0.24–0.92, p = 0.028), in-hospital HF (HR = 3.50, 95% CI: 1.43–8.58, p = 0.006), and periprocedural myonecrosis (HR = 3.20, 95% CI: 1.34–7.67, p = 0.009) were independent predictors of readmission due to HF during 180-day follow up (Table 5).