Introduction

Percutaneous coronary intervention (PCI) for coronary artery disease is a non-surgical invasive procedure with the goal of relieving the narrowing or occlusion of the coronary artery and improving the blood supply to the ischemic tissue [1]. The most common method of PCI consists of ballooning the stenosis lesion or deploying a stent. The CorPath GRX System (Corindus Inc., Waltham, USA), a second-generation robotic system assisting PCI, was first clinically used in the USA, where it was developed, and it demonstrated good treatment outcomes in 40 subjects (54 lesions) [2]. Subsequently, the CorPath GRX, its successor, was introduced in Asia [3, 4] and has been used in Japan since 2019 [5, 6]. Robotic-assisted PCI (R-PCI) has the advantage of lower radiation exposure (95%) to the main operator compared to conventional manual PCI [7]. It has also been reported that precise robotic lesion length measurement contributes to appropriate stent length selection [8] [9]. CorPath GRX is also useful for tortuous and complex lesions [10], and its use in carotid artery disease has recently been reported [11, 12].

Tejas et al. performed 310 R-PCIs using the CorPath GRX from 2017 to 2019 in India and reported significantly lower radiation exposure to the patient without increasing the fluoroscopy time and contrast volume. However, the procedural time was slightly increased (5–10 min) when compared with conventional PCI [3]. In addition, in the study, with respect to 30-day major adverse cardiac or cerebrovascular events (MACCEs), there were eight (2.6%) target lesion restenosis (TLR) and one (0.3%) myocardial infarction, but there was no death and no significant difference from the conventional PCI group. On the other hand, most R-PCIs have been performed under angiography guidance, and there has been scant research on the treatment outcomes of the combination of intravascular imaging.

Intravascular imaging-guided PCI is useful for determining the position of the major axis during stenting and is said to have better treatment outcomes than angiography-guided PCI. Therefore, intravascular imaging-guided R-PCI may demonstrate treatment outcomes better than those of conventional reports. This research is an exploratory study of the treatment outcomes of intravascular imaging-guided R-PCI at a single center, and it aimed to clarify the details of the procedure, clinical outcomes, and procedural complications.

Methods

Study population

The study cohort comprised all 102 consecutive patients (78 males, mean age of 70.1 ± 9.6 years, 110 PCIs, 125 lesions) who underwent R-PCI using the CorPath GRX for ischemic heart disease at Iwate Medical University Hospital from June 12, 2019 to February 18, 2021. A flowchart for inclusion and exclusion criteria are shown in (Fig. 1). Patients characterized with “Acute myocardial infarction,” “culprit lesion in left main trunk,” “culprit lesion in coronary bypass graft,” “with transcatheter aortic valve replacement,” “using debulking device,” “using mechanical circulatory support,” and “without intravascular imaging” were excluded from this study. Of the remaining 401 cases, 110 were selected for R-PCI; the selection of either R-PCI or traditional manual PCI was at the discretion of the main operator.

Fig. 1
figure 1

Flowchart for inclusion and exclusion criteria. Exclusion criteria were “acute myocardial infarction,” “culprit lesion in left main trunk,” “culprit lesion in coronary bypass graft,” “with transcatheter aortic valve replacement,” “using debulking device,” “using mechanical circulatory support,” and “without intravascular imaging”

Full size image

This study was conducted in compliance with the ethical principles of the Declaration of Helsinki (2013, Brazilian revision) and was approved by the ethics committee of Iwate Medical University (MH2021-082). Written informed consent for R-PCI was obtained from all patients, and all patients were given the opportunity to opt out from this retrospective observational study (https://iwate-heart.jp/public_information/).

With regard to the subjects’ clinical characteristics upon admission, all comorbidities and medical history were defined in accordance with the Japanese PCI (J-PCI) registry [13]. Current smoking history was defined as smoking within one year before admission. Left ventricular ejection fraction (LVEF) was measured using the Simpson method by transthoracic echocardiography. The inclusion criteria of this study were (1) stable effort angina pectoris, (2) silent myocardial ischemia, and (3) unstable angina pectoris. PCI was indicated for the stable coronary artery diseases of (1) and (2) when either of the following requirements was met: (1) ≥ 90% stenotic lesion, (2) stenotic lesion possibly causing chest symptoms, and (3) positive for functional ischemia assessments (exercise stress electrocardiogram, stress myocardial scintigraphy, fractional flow reserve, and fractional flow reserve derived from coronary CT angiography. The exclusion criteria were acute myocardial infarction, severe calcified lesion requiring debulking device, and left main trunk lesion.

Procedures

R-PCI was operated mainly by five operators, including two Japanese Association of Cardiovascular Intervention and Therapeutics (CVIT) specialists and three CVIT-certified physicians who received training for the CorPath GRX® system in advance. Meanwhile, setups, including sheath placement, guiding catheter cannulation, and single-use cassette connection, were mainly performed by the assistants.

Procedural time was defined as the time from cannulation to the removal of the guiding catheters used in R-PCI. Intraprocedural iodine contrast volume, patient entrance skin dose, fluoroscopy time, and radiation exposure to the main operator were investigated. All R-PCIs in this study were performed under intravascular imaging (intravascular ultrasound and optical coherence tomography) guidance. The CorPath GRX does not have the function of operating intravascular imaging devices; thus, these devices were used by manual operation. Intravascular imaging was performed at least twice, before and after stenting (or after planned additional dilatation). Before obtaining intravascular imaging, dilators such as isosorbide dinitrate or nicorandil were administered via coronary injection. The size and length of the stents to be implanted were determined by the operators based on the expert consensus document of the European Society of Cardiology [14]. Procedure completion was also judged by each operator based on previously established expert consensus document [14]. In this study, procedural success was defined as cases in which target stenting or balloon dilatation was successful. Procedure details of cases requiring manual operation apart from intravascular imaging (such as guide wire, device, and guiding catheter) were also investigated.

Study end point

The primary end point was the 30-day survival after PCI, while the secondary end points were the PCI complications (myocardial infarction, coronary bypass grafting, stent thrombosis, embolism, hemorrhagic complication, aortic dissection, clear angiographical coronary dissection, pseudoaneurysm, arteriovenous fistula, and renal dysfunction) [15] and cerebrovascular events (myocardial infarction, stroke, and target lesion revascularization) within 30 days. Procedural success was defined as residual stenosis determined by quantitative coronary angiography of < 30%, and final Thrombolysis in Myocardial Infarction (TIMI) flow grade of 3. Based on the Society for Cardiovascular Angiography and Interventions (SCAI) definition, myocardial infarction as a PCI complication was defined as (1) serum creatine kinase-myocardial band (CK–MB) level (reference value 25I U/L) ≥ 10 times of the normal upper limit or myocardial troponin level ≥ 70 times of the normal upper limit or (2) serum CK–MB level ≥ 5 times of the normal upper limit or myocardial troponin level ≥ 35 times of the normal upper limit and the occurrence of a new Q wave or left bundle branch block [16]. Renal dysfunction as a PCI complication was defined as post-PCI serum creatinine level increase of ≥ 2.0 mg/dL or an increase of ≥ 50% from the previous test or new dialysis. Contrast-associated acute kidney injury (AKI) was defined as an increase in the plasma creatinine level to at least 0.5 mg/dL deciliter or at least a 25% increase from the baseline level within 2–5 days after exposure to the contrast [17].

Statistical analysis

All values were expressed as mean ± standard deviation or median, and the Mann–Whitney U test was used for intergroup comparison. SPSS® 25.0 for Windows (IBM, Chicago, U.S.A.) was used for statistical analysis, and p < 0.05 was defined as significantly different for all values.

Results

R-PCIs comprised 15.9% (110 R-PCIs: 106 elective R-PCIs, 4 emergency R-PCIs) of the 693 PCIs (443 elective, 250 emergency) performed at our hospital during the study period. Table 1 shows the subjects’ clinical characteristics and coronary angiography findings. The mean LVEF of this cohort was good at 60.6%. However, left ventricular dysfunction with LVEF < 60% comprised 29.5% of the total, and left ventricular dysfunction with LVEF < 45% comprised 8.6% of the total. Approximately half had a history of ischemic heart disease. Among staged R-PCIs for residual stenotic lesion in patients with acute coronary syndrome who had multivessel disease (n = 29), 72.4% were performed within 30 days after myocardial infarction. Approximately 60% of lesions were of American College of Cardiology Foundation/American Heart Association classification type B2 or C, and chronic total occlusion (CTO) of TIMI flow grade I was confirmed in 2.7% of the total patients (n = 3, Case No. 46, 79, and 99).

Table 1 Clinical characteristics and coronary angiography findings
Full size table

Post-R-PCI 30-day follow-up rate was 100%, and there was no death, cerebrovascular event, or TLR (Table 2). In addition, in this cohort, despite mild to moderate complications, such as subcutaneous bleeding at the puncture site, distal branch coronary artery dissection due to the manual operation of the guide wires, femoral artery pseudoaneurysm, and contrast-associated AKI, there was no procedural complication caused by robotic operation.

Table 2 Complications and clinical outcomes of robotic-assisted percutaneous coronary intervention
Full size table

Table 3 and Fig. 2 show the results of the R-PCI procedure. Intravascular imaging devices were used in all cases. In 41.5% of the cases, additional balloon dilatation was performed in the implanted stents. No case required unplanned additional stenting because of coronary artery dissection. Procedural success for lesions was achieved in all 110 cases that underwent R-PCI. Most R-PCIs had a procedural time of 30–90 min, while cases completed in < 30 min and those requiring > 90 min were approximately 10% each. Based on the graph, procedural time tended to be long in the 13 cases at the early stage of introducing R-PCI, but it reduced significantly afterward (p = 0.003, Mann–Whitney U test, Supplemental Fig. 1). Furthermore, 86.4% of all R-PCIs had a fluoroscopy time of < 30 min, 81.8% had an entrance skin dose of < 1.00 Gy, 83.6% had 0 μSv radiation exposure to the main operator, and 86.4% required iodine contrast volumes within 100 mL. Most of the intravascular imaging modalities used in this study were intravascular ultra sound. Optical coherence tomography was used for patients with severe calcification in the culprit lesion or in-stent restenosis. Intravascular imaging was performed before first ballooning in about half of the cases, and post-stenting in all cases. Touch-up ballooning was added after intravascular imaging in about 40% of post-stenting cases. On the other hand, there was no case that required additional stenting. In all cases, final TIMI flow grade was 3, and percent diameter stenosis after R-PCI was under 30%.

Table 3 Result of robotic-assisted percutaneous coronary intervention
Full size table
Fig. 2
figure 2

Graphs of each factor. Graphs of each factor in (Table 2). It was observed that 77.3% of all R-PCIs had a procedural time of < 60 min, 86.4% had a fluoroscopy time of < 30 min, 81.8% had an entrance skin dose of < 1.00 Gy, 83.6% had 0 μSv radiation exposure to the main operator, and 86.4% had an iodine contrast volume of < 100 mL

Full size image

Manual operation was used in combination in 14 cases (Fig. 3). The procedural time of the robotic operation only group was significantly shorter than that of the manual operation combination group (median 43.5 min vs. 77.5 min, p < 0.001, Mann–Whitney U test, Supplemental Fig. 1). With respect to the details of manual operations, guide wire operation, microcatheters, and distal protection devices were often used for difficult lesions. The switch to manual operation was due to system error in Case No. 52 and single-use cassette error in Case No. 101, but the CorPath GRX operated without any issue after rebooting.

Fig. 3
figure 3

Major causes and details of manual operations besides intravascular imaging device. Manual operations used in combination in addition to intravascular imaging were mostly related to guide wire operations. System error was recovered without any issue following rebooting after R-PCI was completed. Single-use cassette failure was solved after cleaning the connection part of the robotic drive of the adhered contrast agent and rebooting

Full size image

Discussion

This study is the first report on detailed real-world data of intravascular imaging-guided R-PCI performed on > 100 patients. The procedure was safely executed in all cases, the 30-day outcomes were satisfactory, and there were few procedural complications. Based on the results of this study, R-PCI using intravascular imaging could be safe and feasible.

In spite of fluoroscopy and procedural time in this study were longer compared to a previous study [3], the results of the 30-day MACCEs were excellent. Although it is estimated that this difference in outcomes could largely be due to the difference in population, optimal stenting by using intravascular imaging might have contributed to the favorable prognosis [18, 19]. In this cohort, unplanned additional stenting was unnecessary because of the implantation of appropriately sized stents at the appropriate area of culprit lesions. Moreover, there was no 30-day TLR possibly because appropriate balloon dilatation was added by post-stenting intravascular imaging.

In this study cohort, an experience of 13 cases was required to achieve a short R-PCI procedural time. Later, the procedural time was often extended because of the combination of manual operation, challenging cases, or the participation of new operators; nevertheless, the procedural time was generally stable. Therefore, it was estimated that 10–15 cases were required for multiple operators at a single center to become familiarized with the operation of R-PCI, including the operation of intravascular imaging devices via single-use cassettes.

The combination rate of manual operation in this study was similar to that of previous reports [6, 20], and the use of imaging catheter devices did not increase the number of cases that required switching to other manual operations. Most manual operations used in combination involved guide wires or microcatheters. Another center has also reported that the majority of manual operations consisted of microcatheter use [6]. The “technIQ™ Smart Procedural Automation” has been recently developed as a promising new set of automated robotic movements for the CorPath® GRX system for guide wire operation and device delivery. In addition, opportunities for R-PCI utilization will be further expanded when intravascular imaging device operation by robotic arms as well as microcatheter and guide extension catheter use becomes possible in the future.

Based on the good treatment outcomes, we have some expectations for the future of R-PCI, namely R-PCI for complex lesions, such as CTO and severe calcification, and remote PCI. With respect to CTO-PCI, the partially robotic-assisted group had total surgery time, contrast volume, and MACCE frequency equivalent to those of the totally manual PCI group [21]. As CTO-PCI has a long procedural time, the continuation of procedure using robotic assistance after manually wire crossing the lesions by guide wires under microcatheters is a logical strategy to reduce the physical burden and radiation exposure of the operator. Intravascular imaging is effective in PCI for CTO [22]. Good outcomes were achieved in the three cases of CTO that underwent PCI in this study, and we hope to further utilize R-PCI and accumulate more findings in the future. It has been reported that remote PCI has already successfully treated five cases in India [23]. The ability to use intravascular imaging guidance in remote PCI may lead to improvement in the treatment outcomes. Furthermore, the introduction of unprecedented novel technologies, such as the development of imaging modalities that can be combined with R-PCI and the system where artificial intelligence semi-automatically performs stenting based on imaging analysis results, is anticipated.

Study limitations

This study has several limitations. First, this study is a single-center study, and it is possible that subject selection bias might have affected the results. However, the completion of the treatment for > 100 patients without major complications demonstrates the effectiveness of intravascular imaging-guided R-PCI. Second, this cohort is a single-group retrospective observational study of limited cases. Hence, the number of cases and the follow-up period are still insufficient for a precise comparison with conventional manual PCI using intravascular imaging guidance. To overcome this limitation, it is necessary to investigate it further in a large-scale randomized controlled trial. In the future, we plan to accumulate more procedural experience and continue following up on the study cohort.

In conclusion, intravascular imaging-guided R-PCI showed good initial treatment outcomes, and at present, is being safely executed without complications due to robotic maneuver.