ADVERTISEMENT
Outcomes of Robotically Assisted Versus Manual Percutaneous Coronary Intervention: A Systematic Review and Meta-Analysis
Abstract: Objective. We performed a systematic review and meta-analysis of existing studies from the literature comparing robotically assisted (RA) percutaneous coronary intervention (PCI) to manual PCI (M-PCI). Background. RA-PCI is a novel technology that allows the operator to perform PCI from a shielded cockpit using a remote-control module. Methods. MEDLINE/PubMed, EMBASE, and Google Scholar were queried from inception until May 31, 2018 for relevant studies comparing clinical outcomes between RA-PCI and M-PCI. The random-effects model was utilized to compute the summary effect size. Results. Of 2050 retrieved citations, five studies were included, with a total of 148 patients in the RA-PCI arms and 493 patients in the M-PCI control arms. Lower operator radiation exposure was observed with RA-PCI compared with M-PCI. There were no statistically significant differences in total stents per case, PCI time, fluoroscopy time, or procedural success rates between the two groups. Conclusions. In carefully selected patients, RA-PCI was associated with reduced operator radiation exposure compared with M-PCI, but there were no significant differences in procedural success rate, patient radiation exposure, contrast dose, or procedure time.
J INVASIVE CARDIOL 2019;31(8):199-203. Epub 2019 May 15.
Key words: ischemic heart disease, percutaneous coronary intervention, robotics
Coronary artery disease (CAD) is a leading cause of morbidity and mortality, with myocardial infarction (MI) occurring in approximately 660,000 Americans yearly.1,2 Percutaneous coronary intervention (PCI) is performed as a revascularization strategy in a variety of patients with stable angina, acute coronary syndromes, and even silent ischemia.3
Since the introduction of the da Vinci Surgical System (Intuitive Surgical) in 2000, the use of robotics in medical procedures has been progressively widespread. Robots for coronary artery intervention that allow the operator to perform robotically assisted PCI (RA-PCI) from a shielded cockpit using joysticks have been developed and are in clinical use.4 Proposed benefits of RA-PCI over traditional manual PCI (M-PCI) include: less operator radiation exposure, improved outcomes from decreased longitudinal geographic miss, reduced contrast dose, and fewer orthopedic injuries for operators.5-7
As with all novel medical technologies, expanded adoption of RA-PCI is tempered by concerns regarding effectiveness, safety, cost, and clinician familiarity.4 We therefore aimed to perform a systematic review and meta-analysis comparing RA-PCI and M-PCI with respect to various clinical outcomes and procedural measures.
Methods
This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement,8 in addition to recommendations by the Cochrane Collaboration9 and Meta-Analysis of Observational Studies in Epidemiology (MOOSE).10 Searches were performed without any language restrictions. This investigation was registered in PROSPERO (Unique ID Number: CRD42018098886).
Study selection and search strategy. Relevant citations were searched in MEDLINE/PubMed, EMBASE, and Google Scholar, with MeSH strategy and search queries related to hospitalized patients undergoing RA-PCI. All databases were screened from inception until May 31, 2018. References of relevant records were also screened.
Two independent reviewers (JA and DH) screened citations by title and abstract, with discrepancies resolved by consensus. A third reviewer (MA) was available for adjudication. Records of possible relevance to the review were appraised in their entirety if they met selection criteria. Studies were included if they directly compared RA-PCI with M-PCI outcomes. Major exclusion criteria were: (1) duplicate reporting of patient data registries; (2) case reports or series; (3) experimental studies; and (4) lack of M-PCI control arm.
Data extraction and quality assessment. The following data from studies meeting the selection criteria were extracted: journal/conference; authors; year of publication; and location of study. Clinical endpoints of interest were: procedural success rate; contrast media use; fluoroscopy time; procedure time; operator radiation dose; patient dose-area product; longitudinal geographic miss (LGM); and total stents utilized per case. The reviewers assessed the quality of the included studies using a modified version of the Newcastle-Ottawa scale (>7 indicates high quality). Parameters assessed included sample representativeness and size. Funnel plots could not be meaningfully constructed to assess publication bias given the small number of studies.
Statistical analysis. Continuous variables are described as mean ± standard deviation or median (range). Categorical variables are reported as number/total (%). The summary effect size was computed using the random-effects model utilizing MetaXL 5.3 software (EpiGear International). Pooled risk estimates were reported along with their 95% confidence intervals (CIs).
Results
A total of 2050 records were initially retrieved, with 490 duplicate citations (Figure 1). Ultimately, a total of 38 records were deemed worthy of further review upon exclusion of records based on title and abstract appraisal. References of these potentially relevant citations were checked for new records of possible potential interest, and 10 additional citations were found. The full text of 48 citations was reviewed, with 42 studies excluded for reasons outlined in Figure 1. The final analysis therefore included 5 studies that included a total of 148 patients in the RA-PCI arms and 493 patients in the M-PCI control arms. 5,6,11,14,15 All included studies were independently judged to be of high quality by the modified Newcastle-Ottawa scale. The studies are described in Table 1.
Pooled results of meta-analyses describing procedural and clinical outcomes with RA-PCI are shown in Figure 2A-2F. Lower operator radiation exposure was observed with RA-PCI compared with M-PCI (weighted mean difference [WMD] = -15.61 µSv; 95% CI, -30.5 to -0.71) (Figure 2D).
Comparing the RA-PCI and M-PCI arms, there were no statistically significant differences in total stents per case (WMD = 0.0; 95% CI, –0.15 to 0.15) (Figure 2A), contrast media dose (WMD, 3.67 mL; 95% CI, -34.1 to 26.8) (Figure 2B), PCI time (WMD = 5.85 minutes; 95% CI, -1.10 to 12.8) (Figure 2C), fluoroscopy time (WMD = 0.347 minutes; 95% CI, -1.5 to 0.82) (Figure 2E), or clinical procedural success rates (relative risk = 1.0; 95% CI, 0.87 to 1.16) (Figure 2F).
Discussion
We performed a systematic review and meta-analysis evaluating known data comparing RA-PCI with M-PCI in clinical practice. To our knowledge, this represents the first meta-analysis of RA-PCI efficacy. Overall, we found limited data sets directly comparing RA-PCI with M-PCI, given that the majority of studies in this domain are single-arm designs. However, a few trends have emerged to date, namely with regard to radiation exposure, procedure success rates, and procedure times.
With regard to radiation exposure, a proposed benefit of RA-PCI compared with M-PCI is reduced radiation exposure to the cardiologist.16 This is expected, as the cardiologist sits in a radiation-shielded cockpit, and is no longer reliant upon lead garments. Our analysis confirmed a lower operator radiation exposure (Figure 2D), as would be expected with RA-PCI compared with M-PCI. This suggests that RA-PCI may serve as a means to reduce cumulative lifetime radiation exposure, with its associated risks for providers. We did not encounter any long-term studies of providers with regard to sequelae of radiation exposure or orthopedic injury related to lead garments; this is not an unexpected gap, given the relatively recent introduction of RA-PCI. These benefits are based on the linear no-threshold model; however, current evidence supports the underlying premise of reduced radiation exposure in the absence of heavy lead garments. As this robotic technology advances, RA-PCI may be performed remotely, and telerobotics may come to hold a place in the toolset of the practicing interventional cardiologist.4
It has been postulated that while RA-PCI might reduce operator radiation exposure, it may increase patient exposure, by a variety of possible mechanisms, such as increased procedure time or decreased provider prudence with fluoroscopy.15 However, there was no difference found in patient radiation dose between RA-PCI and M-PCI (Figure 2D), with the evidence to date suggesting that reduced provider radiation exposure does not come at a cost of increased patient exposure.
No difference in use of contrast media was noted between the two arms (Figure 2B), failing to substantiate the hypothesis that RA-PCI may reduce nephrotoxic contrast dose.
No difference was found in clinical success rates (defined as <30% residual stenosis of stented lesions without in-hospital major adverse cardiovascular events (Figure 2F) between RA-PCI and M-PCI comparison groups. One limitation of this analysis is the paucity of long-term clinical outcome data in any of the included studies. Nearly universal success rates were seen in both arms; no statement of superiority of either RA-PCI or M-PCI can be made based on the available data.
With the introduction of any procedural technology, effects on procedural success rate and time of procedure are of particular interest. Neither total PCI time nor fluoroscopy time was found to be significantly different between the RA-PCI and M-PCI arms (Figure 2E). The included studies were mixed with regard to provider experience, which may underestimate learning-curve effects for providers inexperienced with RA-PCI. Notably, Kohan et al examined effects of the learning curve in a single new cardiology fellow, and showed no difference between the procedure times in the first 15 cases of the RA-PCI and M-PCI groups.17
Study limitations. For certain variables, the number of studies eligible for analysis were few, with the subsequent limitation the total patients included. Few large, controlled, high-quality studies have been completed to date. This may statistically obscure true differences that may exist. Further research is needed given the limited number of data sets, as the results reflect carefully selected studies. However, even given the relatively small data sets, statistically decreased provider radiation exposure was noted. Additionally, it remains possible that there are additional data sets that were not found despite our systematic approach to data acquisition.
Finally, two significant data sets were excluded due to lack of M-PCI control arm. The PRECISE study is a prospective, single-arm, open-label study of the CorPath 200 RA-PCI system that enrolled 163 patients.18 Bezerra et al compared this data set to a reference population and found a significantly lower longitudinal geographic miss rate.19 However, we did not include this study in the final analysis given the lack of a true M-PCI control arm; only a subset of 40 patients was included, as it was compared against a control arm.18
Additionally, the ongoing PRECISION trial is a single-arm, open-label, multicenter, post-market registry of the CorPath 200 System20 that was not included in the present analysis due to the lack of M-PCI control arm. Preliminary results for 754 patients and 949 lesions were presented at the 2017 Society for Cardiovascular Angiography and Interventions meeting, showing technical success rates of 88.6% for radial vs 82.4% for transfemoral, with technical success defined as <30% residual stenosis, TIMI 3 flow, no conversion to manual PCI, no major adverse cardiovascular events. In addition, rates of clinical success, defined as procedural success without major adverse cardiovascular events were 98.9% for transradial vs 94.9% for transfemoral approach.21 A subset of 108 RA-PCI procedures in the PRECISION registry was included in the analysis, as it was reported separately with a manual control group.14
Conclusion
In carefully selected patients, a systemic meta-analysis of studies directly comparing RA-PCI with M-PCI suggests that RA-PCI leads to less operator radiation exposure compared with M-PCI, with no significant differences in procedural success.
References
1. Dalen JE, Alpert JS, Goldberg RJ, Weinstein RS. The epidemic of the 20(th) century: coronary heart disease. Am J Med. 2014;127:807-812.
2. Mozaffarian D, Benjamin EJ, Go AS, et al. Executive summary: heart disease and stroke statistics – 2016 update: a report from the American Heart Association. Circulation. 2016;133:447-454.
3. Torpy JM, Lynm C, Glass RM. Percutaneous coronary intervention. JAMA. 2004;291:778.
4. Maor E, Eleid MF, Gulati R, Lerman A, Sandhu GS. Current and future use of robotic devices to perform percutaneous coronary interventions: a review. J Am Heart Assoc. 2017;6:e006239.
5. Madder RD, VanOosterhout S, Mulder A, et al. Impact of robotics and a suspended lead suit on physician radiation exposure during percutaneous coronary intervention. Cardiovasc Revasc Med. 2017;18:190-196.
6. Campbell PT, Kruse KR, Kroll CR, Patterson JY, Esposito MJ. The impact of precise robotic lesion length measurement on stent length selection: ramifications for stent savings. Cardiovasc Revasc Med. 2015;16:348-350.
7. Mahmud E, Pourdjabbar A, Ang L, Behnamfar O, Patel MP, Reeves RR. Robotic technology in interventional cardiology: current status and future perspectives. Catheter Cardiovasc Interv. 2017;90:956-962.
8. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535.
9. Higgins JPT GS. Cochrane Handbook for Systematic Reviews of Interventions. 5.2. West Sussex: John Wilesy; 2017. https://training.cochrane.org/handbook. Accessed May 28, 2018.
10. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA. 2000;283:2008-2012.
11. Beyar R, Gruberg L, Deleanu D, et al. Remote-control percutaneous coronary interventions. J Am Coll Cardiol. 2006;47:296-300.
12. Campbell P, Tennis P, Bitler C, et al. Staff exposure to X-ray during PCI: randomized comparison of robotic vs manual procedures. Catheter Cardiovasc Interv. 2016;87:S80-S81.
13. Madder RD, VanOosterhout S, Mulder A, et al. Impact of robotics and a suspended lead suit on physician radiation exposure during percutaneous coronary intervention. Cardiovasc Revasc Med. 2017;18:190-196.
14. Mahmud E, Naghi J, Ang L, et al. Demonstration of the safety and feasibility of robotically assisted percutaneous coronary intervention in complex coronary lesions: results of the CORA-PCI study (Complex Robotically Assisted Percutaneous Coronary Intervention). JACC Cardiovasc Interv. 2017;10:1320-1327.
15. Smilowitz NR, Moses JW, Sosa FA, et al. Robotic-enhanced PCI compared to the traditional manual approach. J Invasive Cardiol. 2014;26:318-321.
16. Naghi J, Harrison J, Ang L, et al. Does robotic percutaneous coronary intervention lower stent use compared to the manual approach? Catheter Cardiovasc Interv. 2016;87:S90-S91.
17. Kohan L, Nagarajan V, Ragosta M. Robot vs human: effect of a robot-assisted PCI system on procedure times. Catheter Cardiovasc Interv. 2016;87:S86.
18. Weisz G, Metzger C, Caputo R, et al. Corpath-PRECISE: final results of the first pivotal study for robotically-enhanced PCI. J Am Coll Cardiol. 2012;59:E299.
19. Bezerra HG, Mehanna E, Vetrovec G, Costa M, Weisz G. Longitudinal geographic miss (LGM) in robotic assisted versus manual percutaneous coronary interventions. J Interv Cardiol. 2015;28:449-455.
20. U.S. National Library of Medicine. PRECISION GRX Registry. Accessed at https://clinicaltrials.gov/ct2/show/NCT03278301.
21. Large Multicenter Study Shows High Success Rate for Robotic PCI Procedures [press release]. May 12, 2017. Available at https://www.scai.org/Press/detail.aspx?cid=eb588694-bc51-4552-b8c9-6b50c6daff87#.XMhxTqZ7mfU
*Joint first authors
From the Section of Cardiology, Baylor College of Medicine, Houston, Texas.
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.
Manuscript submitted January 9, 2019, provisional acceptance given January 25, 2019, final version accepted January 31, 2019.
Address for correspondence: Joseph Allencherril, MD, Department of Medicine, Baylor College of Medicine, 1 Baylor Plaza, Houston, TX 77030. Email: jallencherril@gmail.com
