Radial access for percutaneous coronary intervention (PCI) has been adopted by many interventional cardiologists as an alternative vascular access site during the last decade. The transradial approach is the preferred access strategy for PCI in several European countries, Canada, and Japan, with momentum building more slowly in the United States.1-4 Previous studies have demonstrated reduced length of stay (LOS),5,6 fewer access site complications,4,7 improved quality of life and patient satisfaction,5,6 and earlier ambulation6 for transradial PCI (TRI) when compared to transfemoral PCI (TFI).

These well-documented clinical benefits, in addition to reduced LOS, transform TRI into an economically compelling strategy.5,7-10 However, these clinical benefits for TRI occur at the cost of increased access site crossover11,12 and reduced procedural success,13 potentially resulting in increased resource utilization.8,11,12 Increased operator expertise during a defined learning curve6,14,15 improves the success rate and procedure duration for TRI.16

The cardiovascular services line is under tremendous pressure to provide cost-effective treatment while maintaining quality of care. Physicians and hospital administrators struggle with understanding the potential economic benefits that TRI might offer compared to TFI.8,17 To better understand the economic impact of TRI, we compared the costs and quality outcomes (bleeding, LOS, and mortality) between TRI and TFI based on experience in a large-volume tertiary care center.

METHODS

Study Design

Our study retrospectively compared the cost and clinical outcomes between TRI and TFI at the Frederik Meijer Heart & Vascular Institute, Spectrum Health in Grand Rapids, Michigan. Spectrum Health–Butterworth Hospital is a 989-bed teaching hospital in West Michigan and is the primary teaching hospital for Michigan State University–College of Human Medicine. The study was approved by the hospital's institutional review board. Twenty-one experienced interventional cardiologists were involved in performing TFI or TRI PCI during the course of the study.

All patients who underwent PCI from January 2010 through March 2011 were included. The exclusion criteria included patients who had more than one PCI during the same hospital stay. The primary outcome measures were adjusted total, procedural, and postprocedural costs, which were used to determine cost differences between TRI and TFI. The secondary objective was to evaluate quality outcomes: post-PCI bleeding within 72 hours, LOS, and allcause in-hospital mortality for TRI and TFI.

Study Population and Data Collection

A total of 2,972 patients were extracted from the hospital data warehouse (SAP Business Objects–Enterprise XI, Product: 12.1.0, Walldorf, Germany) using ICD-9 (International Classification of Disease–version 9) principal and secondary procedure code 0.66. Fifty-three patients were identified as having undergone repeat PCI during the same hospital stay, resulting in a final cohort of 2,919 patients. The data components with their element numbers were obtained from the National Cardiovascular Data Registry (NCDR) CathPCI Registry.

The demographic data (with their element numbers) collected included age, sex (2060), race (2070–2074), New York Heart Association class (5040, 5045), Canadian Cardiovascular Society class (5020), creatinine level (7315), hypertension (4005), coronary artery disease presentation (5000) (ST-elevation myocardial infarction [STEMI], non-STEMI, unstable angina), cardiogenic shock (5060), history of congestive heart failure (4025), PCI (4035), and peripheral vascular disease (4075). In addition, primary insurance type (3020–3027) was collected.

The procedural information included anticoagulant therapy (9500, 9510), access location (femoral, radial) (5350), closure device (5355), number and type of stents (7225), PCI status (7020) and indication (7035), coronary artery bypass graft (9000, 9005, 9010, 9015), and inpatient or outpatient procedure (9065). The outcomes data included bleeding events (8050, 8055, 8060, 8061, 8070, 8080, 8090, 8100), in-hospital mortality (9040, 9055), and total LOS (discharge date [9035] minus admit date [3000]).

The economic analysis probed the total cost (cost on the day of PCI through hospital discharge), procedural cost, and postprocedural cost to assess if cost differences were being determined by different stages in a hospitalization. The costs were calculated for the day of procedure (procedural costs) and the day after the procedure through discharge (postprocedural costs) based on the billing day for each item description. As part of the subanalysis, total costs alone were reported for same-day discharge patients, as they did not have postprocedural costs. All cost components were an aggregate of direct and indirect costs. The cost data were linked to the NCDR CathPCI Registry data to form a combined deidentified dataset. The study data were sent to the third-party Vita Solutions firm (a subsidiary of The Medicines Company, Parsippany, NJ) that performed data summary and statistical analysis.

Statistical Analysis

Continuous variables were expressed as mean ± standard deviations, and categorical variables were expressed as counts with percentages. Intergroup differences for continuous variables were tested using Student's t-test. The chi-square test was used to determine the differences in categorical variables between the TRI and TFI. Due to outliers in cost data, all costs were trimmed back to the 95% confidence level of the mean. A generalized linear mixed model18 was developed for total cost, procedural cost, and postprocedural cost. The nine covariates in the model included PCI status, race, sex, previous congestive heart failure, peripheral vascular disease, previous PCI, New York Heart Association class, cardiogenic shock, and angina type at admission. The bootstrap method (1,000 repetitions) with replacement was applied to account for skewness in the data19,20 and to determine the cost difference and 95% confidence intervals (CIs).

The secondary endpoint was to determine if there were differences in LOS between TRI and TFI. These differences were estimated using a generalized linear mixed model with the covariates. Additional endpoints (in-hospital bleeding and mortality) were analyzed using unadjusted logistic regression and odds ratios calculated when appropriate. The population was further stratified into low-, moderate-, and high-bleeding-risk categories according to the NCDR CathPCI bleeding risk model,21,22 with total, procedural, and postprocedural costs analyzed for each subgroup. Statistical significance was defined as P ≤ .05, and all statistical analyses were performed using SAS Version 9.2 (SAS Institute, Cary, NC).

RESULTS

Demographics

Of the total study population (n = 2,919), 191 patients (6.5%) underwent TRI. Baseline clinical characteristics are summarized in Table 1. TRI patients were younger, less likely to undergo primary PCI for STEMI, had lower prevalence of Canadian Cardiovascular Society class IV, and lower probabilities of bleeding risk. Additional differences are notable in the anticoagulation strategies and closure devices used. TRI patients had more frequent use of unfractionated heparin (with or without glycoprotein IIb/IIIa inhibitors) and less frequent use of bivalirudin. The difference in closure device usage was expected because it was not required in the TRI cohort.

Economic Outcomes

The study cohort consisted of 961 patients with low bleeding risk, 1,413 with moderate risk, and 545 with high risk. The unadjusted and adjusted cost savings by bleeding risk category are shown in Table 2. The total unadjusted costs were $1,975 lower in the TRI than the TFI group (95% CI, $944–$3,007; P < .001). After bleeding risk adjustment, the difference decreased to $1,375 (95% CI, $531–$2,056; P = .008). TRI patients also had statistically significant total adjusted cost savings in the low-risk and moderate-risk subgroups. In the high-bleeding-risk category, both adjusted and unadjusted costs were similar between TFI and TRI.

The unadjusted procedural cost savings for TRI was $1,019. The adjusted and unadjusted procedural costs were statistically significant in all cases and in moderate-risk subgroups, with a cost difference favoring TRI. The unadjusted postprocedural cost savings for TRI was $957 (95% CI, $119– $1,794; P = .025). After adjustment, the TRI group did not achieve statistically significant postprocedural cost savings. The adjusted postprocedural cost savings achieved statistical significance in the low-risk subgroup analyses, favoring TRI.

Same-Day Discharge

Utilizing our institutional same-day discharge (SDD) guidelines,23 5.9% of the study cohort was discharged on the same day after low-risk elective PCI. As shown in Table 1, SDD was statistically more likely to occur in the TRI cohort (19.9% vs 4.9%; P < .001). TRI was associated with a statistically significant unadjusted and adjusted total cost savings of $1,126 and $1,064, respectively, in the SDD population (Table 3).

Quality Outcomes

The unadjusted quality outcomes are summarized in Table 4. Overall, bleeding events (within 72 hours) in the TRI and TFI groups totaled 1.6% and 3.1%, respectively. The unadjusted bleeding events in TRI and TFI groups were similar (P > .05) for the entire study cohort and among different bleeding risk strata. As shown in Figure 1, the adjusted odds ratio was not statistically significant in all cases or in the subgroup analyses. Unadjusted in-hospital mortality in TRI and TFI occurred in 1.6% and 1.2% of the cases, respectively (P = .732). In the unadjusted model for in-hospital mortality, the difference was not statistically significant (odds ratio, 1.26; 95% CI, 0.38–4.16). The odds ratios for in-hospital mortality were not calculated by risk category due to the very low event rates in the bleeding risk strata.

The mean LOS for the TRI group was 2.1 days as compared to 2.7 days for the TFI group, a difference of 0.6 days favoring TRI (P = .004). There was a trend toward statistical significance in the moderate-risk category, with a 0.5-day shorter mean LOS (P = .052). The LOS was 0.3 days shorter (P = .104) in the low-bleeding-risk group and 0.8 days shorter (P = .456) in the high-risk group. After adjustment, the differences narrowed but remained statistically insignificant (Table 5).

DISCUSSION

This study provided a comprehensive analysis of the cost savings associated with TRI compared to TFI based on the experience of a single, large-volume tertiary care center. The fact that we were early in our TRI experience is notable, as the learning curve might hinder favorable TRI outcomes. Previous studies on small populations have shown cost savings ranging from $77 to $289 for diagnostic coronary angiography.5,6,9 A large, nationwide administrative database demonstrated a direct hospital cost savings of $553 for TRI relative to TFI.24 Strikingly, our observations showed an average total cost savings of approximately $1,375 per patient for TRI, primarily driven by the 0.6-day shorter LOS. In addition, significant total cost savings were seen in the low- and moderate-bleeding-risk categories. It is important to note that our cost data were an aggregate of direct and indirect costs, which may explain the larger cost savings.

The procedural costs in our population were significantly lower for TRI. Although procedural equipment costs for transradial stent procedures are slightly higher than those for femoral procedures, this could be offset by lower costs for complications.10 We believe this could be the scenario here. Additional explanations include lower supply costs and fewer access site complications for transradial stenting procedures, as previously demonstrated by Mann et al.8

A general consensus has been that cost savings for TRI occur after the procedure, primarily due to lower costs of complications and nursing care. Any decrease in postprocedural complications can potentially result in a reduction in health care expenses. The postprocedural costs in our cohort showed a statistically insignificant trend, yet with some savings, toward TRI. Our observations extend our findings to suggest radial access as a preferred strategy with overall cost savings.

More than 1 million cardiac catheterizations and 500,000 PCIs are performed annually in the United States alone.25 Radial artery cases account for < 10% of the diagnostic cases and approximately 1% of PCI cases.4 TRI has been demonstrated to be safe, with reduced mortality, vascular complications, and major bleeding—as shown in randomized clinical trials of select populations.13,26 Broader adoption of radial catheterization promises to provide these benefits at a reduced cost.

SDD after uncomplicated PCI has been adopted in many centers and is recognized as a safe and effective strategy.18,27-31 In addition, payers have switched elective PCI to outpatient designation and reduced reimbursement.32 A total SDD cost savings of approximately $1,000 per TRI in our analysis is of significant economic consideration for both payers and hospital administrators. In an increasing scenario of SDD PCI, a switch to TRI in low-risk or uncomplicated PCI could significantly decrease the cost burden. The magnitude of overall cost savings with TRI, especially when combined with SDD, should catapult hospitals into trying to implement TRI programs.

Previous studies have shown fewer bleeding complications with TRI as compared to TFI.4,13,17,33 In contrast, our results showed similar bleeding events in both the TRI and TFI groups. The nonrandomized nature of the study may have led to this biased result.

Finally, any attempt to reduce the LOS saves cost. Consistent with multiple other studies,5,6 our results showed reduced LOS for TRI. Of note, 5.9% of the patients were discharged on the same day as the procedure. It is imperative to understand that the TRI cohort's reduced LOS due to early ambulation and fewer vascular complications directly led to savings.

STRENGTHS AND LIMITATIONS

Several limitations of our analysis should be noted. First, these observations reflect the clinical practices of 21 experienced interventional cardiologists at a single United States academic medical center. Of note, only a minority of these interventionists used TRI as the preferred access route. The results may, therefore, not be generalizable to centers with low volumes of PCI, early stages of TRI adoption,14,15,34 or different practice patterns. Second, as a retrospective nonrandomized study, unmeasured confounders may have influenced physician choice of access site and caused selection bias. Our population included high-risk cohorts, such as patients with cardiogenic shock, STEMI, emergent procedures, and coronary artery bypass graft during admission. It is plausible that the TRI population had fewer high-bleeding- risk patients (9.9% vs 19.3%), which would favor more cost savings. In addition, access site crossover data were not collected. There may be a negligible variation in reported rates of TRI and TFI. Third, this study was not powered for clinical outcomes, namely bleeding events and in-hospital mortality, to derive meaningful conclusions.

Our cost data on TRI and TFI across a spectrum of bleeding risk cohorts are based on a validated bleeding prediction model. Thus, it provides detailed cost differences in clinically important subgroups. We included both inpatients and outpatients. It is conceivable that physicians could have discharged TRI cohort patients earlier due to fewer postprocedural complications or SDD preference. This would partially account for the cost differences noted in this study from reduced LOS. Our cost data were captured at the hospital level in a large administrative database and analyzed from the hospital's perspective because costs were calculated based on reports of actual hospital expenditures.

CONCLUSION

In this study, TRI was associated with a total cost savings of $1,375 per patient when compared to TFI, primarily driven by procedural costs savings and shorter LOS. The influence of such savings could provide critical momentum in the shift from a TFI to a TRI approach in PCI.

Acknowledgement: Editorial assistance in the preparation of this article was provided by Beyond Words, Inc. Although Beyond Words supplied paid professional editing services, this does not indicate its endorsement of, agreement with, or responsibility for the content of the article.

Purushothaman Muthusamy, MD, is with the Internal Medicine Residency Program, Michigan State University/ Grand Rapids Medical Education Partners in Grand Rapids, Michigan. He has stated that he has no financial interests related to this article. Dr. Muthusamy may be reached at (616) 732-6232; purush.muthusamy@spectrumhealth.org.

David H. Wohns, MD, FACC, FSCAI, is with the Department of Cardiology, Division of Interventional Cardiology, Frederik Meijer Heart & Vascular Institute; and Division of Cardiology, Spectrum Health in Grand Rapids, Michigan. He has stated that he has no financial interests related to this article.

  1. Bertrand OF, Rao SV, Pancholy S, et al. Transradial approach for coronary angiography and interventions: results of the first international transradial practice survey. JACC Cardiovasc Interv. 2010;3:1022-1031.
  2. Subherwal S, Peterson ED, Dai D, et al. Temporal trends in and factors associated with bleeding complications among patients undergoing percutaneous coronary intervention: a report from the National Cardiovascular Data CathPCI Registry. J Am Coll Cardiol. 2012;59:1861-1869.
  3. Dehmer GJ, Weaver D, Roe MT, et al. A contemporary view of diagnostic cardiac catheterization and percutaneous coronary intervention in the United States: a report from the CathPCI Registry of the National Cardiovascular Data Registry, 2010 through June 2011. J Am Coll Cardiol. 2012;60:2017-2031.
  4. Rao SV, Ou FS, Wang TY, et al. Trends in the prevalence and outcomes of radial and femoral approaches to percutaneous coronary intervention: a report from the National Cardiovascular Data Registry. JACC Cardiovasc Interv. 2008;1:379-386.
  5. Cooper CJ, El-Shiekh RA, Cohen DJ, et al. Effect of transradial access on quality of life and cost of cardiac catheterization: a randomized comparison. Am Heart J. 1999;138(3 Pt 1):430-436.
  6. Louvard Y, Lefevre T, Allain A, Morice M. Coronary angiography through the radial or the femoral approach: the CARAFE study. Catheter Cardiovasc Interv. 2001;52:181-187.
  7. Jolly SS, Amlani S, Hamon M, et al. Radial versus femoral access for coronary angiography or intervention and the impact on major bleeding and ischemic events: a systematic review and meta-analysis of randomized trials. Am Heart J. 2009;157:132-140.
  8. Mann T, Cowper PA, Peterson ED, et al. Transradial coronary stenting: comparison with femoral access closed with an arterial suture device. Catheter Cardiovasc Interv. 2000;49:150-156.
  9. Roussanov O, Wilson SJ, Henley K, et al. Cost-effectiveness of the radial versus femoral artery approach to diagnostic cardiac catheterization. J Invas Cardiol. 2007;19:349-353.
  10. Kiemeneij F, Hofland J, Laarman GJ, et al. Cost comparison between two modes of Palmaz Schatz coronary stent implantation: transradial bare stent technique vs. transfemoral sheath-protected stent technique. Cathet Cardiovasc Diagn. 1995;35:301-308, discussion 309.
  11. Agostoni P, Biondi-Zoccai GG, de Benedictis ML, et al. Radial versus femoral approach for percutaneous coronary diagnostic and interventional procedures: systematic overview and meta-analysis of randomized trials. J Am Coll Cardiol. 2004;44:349-356.
  12. Dehghani P, Mohammad A, Bajaj R, et al. Mechanism and predictors of failed transradial approach for percutaneous coronary interventions. JACC Cardiovasc Interv. 2009;2:1057-1064.
  13. Jolly SS, Yusuf S, Cairns J, et al. Radial versus femoral access for coronary angiography and intervention in patients with acute coronary syndromes (RIVAL): a randomised, parallel group, multicentre trial. Lancet. 2011;377:1409-1420.
  14. Ball WT, Sharieff W, Jolly SS, et al. Characterization of operator learning curve for transradial coronary interventions. Circ Cardiovasc Interv. 2011;4:336-341.
  15. Looi JL, Cave A, El-Jack S. Learning curve in transradial coronary angiography. Am J Cardiol. 2011;108:1092-1095.
  16. Louvard Y, Lefevre T, Morice MC. Radial approach: what about the learning curve? Cathet Cardiovasc Diagn. 1997;42:467-468.
  17. Mann T, Cubeddu G, Bowen J, et al. Stenting in acute coronary syndromes: a comparison of radial versus femoral access sites. J Am Coll Cardiol. 1998;32:572-576.
  18. Blough DK, Madden CW, Hornbrook MC. Modeling risk using generalized linear models. J Health Econ. 1999;18:153-171.
  19. Efron B, Tibshirani RJ. An Introduction to the Bootstrap. New York, NY: Chapman & Hall; 1993.
  20. Mahoney EM, Chu H. Cost-effectiveness analysis alongside clinical trial: statistical and methodological issues. In: Weintraub WS, ed. Contemporary Cardiology: Cardiovascular Health Care Economics. Totowa, NJ: Humana Press Inc; 2003:123-156.
  21. Marso SP, Amin AP, House JA, et al. Association between use of bleeding avoidance strategies and risk of periprocedural bleeding among patients undergoing percutaneous coronary intervention. JAMA. 2010;303:2156-2164.
  22. Mehta SK, Frutkin AD, Lindsey JB, et al. Bleeding in patients undergoing percutaneous coronary intervention: the development of a clinical risk algorithm from the National Cardiovascular Data Registry. Circ Cardiovasc Interv. 2009;2:222-229.
  23. Muthusamy P, Busman DK, Davis AT, Wohns DH. Assessment of clinical outcomes related to early discharge after elective percutaneous coronary intervention: COED PCI. Catheter Cardiovasc Interv. 2013;81:6-13.
  24. Safley DM, Amin AP, House JA, et al. Comparison of costs between transradial and transfemoral percutaneous coronary intervention: a cohort analysis from the Premier research database. Am Heart J. 2013;165:303-309 e2.
  25. Roger VL, Go AS, Lloyd-Jones DM, et al. Heart disease and stroke statistics—2012 update: a report from the American Heart Association. Circulation. 2012;125:e2-e220.
  26. Chase AJ, Fretz EB, Warburton WP, et al. Association of the arterial access site at angioplasty with transfusion and mortality: the M.O.R.T.A.L study (Mortality Benefit of Reduced Transfusion After Percutaneous Coronary Intervention via the Arm or Leg). Heart. 2008;94:1019-1025.
  27. Slagboom T, Kiemeneij F, Laarman GJ, van der Wieken R. Outpatient coronary angioplasty: feasible and safe. Catheter Cardiovasc Interv. 2005;64:421-427.
  28. Ranchord AM, Prasad S, Seneviratne SK, et al. Same-day discharge is feasible and safe in the majority of elderly patients undergoing elective percutaneous coronary intervention. J Invas Cardiol. 2010;22:301-305.
  29. Gilchrist IC, Rhodes DA, Zimmerman HE. A single center experience with same-day transradial-PCI patients: a contrast with published guidelines. Catheter Cardiovasc Interv. 2012;79:583-587.
  30. Patel M, Kim M, Karajgikar R, et al. Outcomes of patients discharged the same day following percutaneous coronary intervention. JACC Cardiovasc Interv. 2010;3:851-858.
  31. Rao SV, Kaltenbach LA, Weintraub WS, et al. Prevalence and outcomes of same-day discharge after elective percutaneous coronary intervention among older patients. JAMA. 2011;306:1461-1467.
  32. Furl N. The movement to outpatient coronary interventions. Cath Lab Digests [serial online]; December 2008.
  33. Hamon M, Rasmussen LH, Manoukian SV, et al. Choice of arterial access site and outcomes in patients with acute coronary syndromes managed with an early invasive strategy: the ACUITY trial. EuroIntervention. 2009;5:115-120.
  34. Leonardi RA, Townsend JC, Bonnema DD, et al. Comparison of percutaneous coronary intervention safety before and during the establishment of a transradial program at a teaching hospital. Am J Cardiol. 2012;109:1154-1159.