Quality improvement initiative to increase radial artery usage as a second arterial conduit in coronary artery bypass grafting

Abstract

OBJECTIVES

Use of radial artery as a second arterial graft, compared to a saphenous vein, in coronary artery bypass grafting (CABG) can improve late outcomes. However, the radial artery remains underutilized. We initiated a quality improvement (QI) initiative to increase the usage of radial artery grafts.

METHODS

During our 4-month lead period, we disseminated evidence for radial artery graft usage to surgeons, developed a radial artery decision-making algorithm and adopted endoscopic harvesting. Our QI initiative was conducted over a 6-month period and included a postoperative survey of decision-making for graft selection and obstacles to radial artery usage.

RESULTS

Over the 6-month study period, 247 patients received isolated CABG which included 98 (40%) with radial arteries as a second arterial graft and 144 (58%) with greater saphenous veins. Radial artery usage increased with QI initiative implementation by 67% compared to 6 months prior to the study period (60 radial arteries/252 isolated CABG, 24%) (P = 0.006). The survey response rate was 93% (231/247). Barriers to radial artery graft usage were poor quality target vessel or stenosis <80% (24%), patient age >75 years (20%), ejection fraction ≤35% (8%) and renal insufficiency/dialysis (7%). No patients experienced significant complications from radial artery harvest.

CONCLUSIONS

Our institutional QI initiative was successful in (i) increasing the usage of radial artery as a second arterial graft and (ii) understanding barriers to radial artery graft usage. Implementation of a QI program can improve radial artery usage in CABG with low risk of patient morbidity from radial artery harvest.

Coronary artery bypass grafting (CABG) is the most common cardiac surgery performed every year worldwide [1].

Graphical Abstract

INTRODUCTION

Coronary artery bypass grafting (CABG) is the most common cardiac surgery performed every year worldwide [1]. Owing to the body of evidence-driven outcomes and its high patency, the internal mammary artery (IMA), most commonly the left (left IMA), is used as the graft conduit to the left anterior descending (LAD) coronary artery [2]. The saphenous vein is the most common second graft choice, with graft failure rates ranging from 10% to 30% within the first year after CABG and an additional 5% each subsequent year after surgery [3–5]. Notably, half of the patients who receive a vein graft will experience graft failure at 10 years post-CABG [6]. Furthermore, graft failure is associated with patient death, myocardial infarction (MI) and need for repeat revascularization [6]. Graft choice is thus critical to optimizing outcomes after CABG.

Conversely, multi-arterial grafting with the radial artery in CABG is associated with improved patency and clinical outcomes in select patient populations in both observational studies and clinical trials [7, 8]. Radial artery grafts demonstrate higher patency rates than saphenous vein grafts [9], with significantly lower incidence of complete graft occlusion and graft failure and higher complete graft patency [10]. One large cohort study demonstrated a difference in graft patency of 15% between the radial artery and saphenous vein graft groups over a median follow-up of 8.6 years [9]. Clinically, radial artery grafts are associated with lower morbidity, including reduction in incidence of MI or repeat revascularization [6, 11, 12]. When compared to the right internal mammary artery (RIMA), radial artery grafts had a lower incidence of deep sternal wound infections and a similar rate when compared to saphenous vein grafts [13, 14]. The 15-year Radial Artery Patency and Clinical Outcomes (RAPCO) randomized trial reported significantly lower rates of death, MI and repeat revascularization with radial artery compared to saphenous vein and RIMA grafts [15]. Finally, the American College of Cardiology/American Heart Association revascularization 2021 guidelines include a class 1 recommendation to use a radial artery as a graft for the second most important target vessel containing significant stenosis in patients with acute coronary syndromes or stable ischaemic cardiac disease [16].

Nevertheless, clinical practice has been slow to adopt the practice of radial artery grafting for CABG. The Society for Thoracic Surgeons (STS) national database reported that, among patients undergoing isolated primary CABG between 2004 and 2015, only 6.5% of patients received a radial artery as a conduit and even fewer (4.9%) received bilateral IMAs [17]. Analysis of recent national trends continues to show underutilization of multi-arterial grafting [18]. A surgeon’s decision on graft choice integrates patient factors, coronary anatomy and surgeon experience [7, 19]. This study aims to implement a quality improvement (QI) initiative to increase evidence-based usage of radial artery grafts as a second conduit at a single academic institution.

METHODS

Ethical statement

This study was approved by the Duke University Health Institutional Review Board (Pro00108176) on April 26, 2021. The need for patient consent was waived.

Eligibility

Consecutive patients undergoing isolated CABG via median sternotomy with 2 or more grafts for severe coronary artery disease, between 1 November 2020 and 30 April 2022, were included. Patients with a single graft or concomitant procedure (i.e., aortic valve replacement) were excluded. The study cohort was compared to patients who underwent isolated CABG 6 months prior to initiation of the QI study period. The comparator group was retrospectively identified from a prospectively maintained institutional database.

Quality improvement initiative

In preparation for the QI initiative, we interviewed surgeons and support staff to understand the baseline institutional practice patterns and workflow contributing to graft choice and harvest. We then executed a 4-month lead period as detailed in Fig. 1. Each month, the cardiac surgeons reviewed updated evidence for radial artery usage in the form of division journal clubs and presentations and ultimately reached consensus on the ‘Radial Artery Decision-making Algorithm’ (Fig. 2). The algorithm was distributed in the clinics and the operating room scheduling office, and the schedulers and physician assistants (PAs) were instructed to inquire about usage of a radial artery graft if the patient met criteria. Preoperatively, the Allen’s test was used to assess for palmar arch patency, which was confirmed intraoperatively [20]. The lead period also included training surgical PAs on endoscopic radial artery harvesting techniques.

Figure 1:

Methodological design of this study showing the elements accomplished in the 4-month lead period and the components of the 6-month QI initiative period.

Figure 2:

Algorithm of evidence-based patient selection for usage of radial artery graft as a second graft in cases of isolated coronary artery bypass grafting.

After the lead period, the QI project was conducted over a 6-month period (1 November 2021–30 April 2022). Cardiac surgeons completed a postoperative survey consisting of 8 questions on graft choice decision-making and operational obstacles. Surgeons were allowed to choose >1 reason that they did not use the radial artery for grafting. The survey was emailed within 8 hours of case completion. If no survey response was received within 24 hours, a reminder was sent to the cardiac surgeon. If no response was received within 48 hours after case completion, the survey was marked as no response received, and the case was excluded from analysis.

The primary objective of our study was to increase evidence-based radial artery graft usage. Secondary objectives were to identify (i) operational obstacles to radial artery use and (ii) complications of radial artery harvest.

Statistical analysis

Patient data and operative details were collected with manual chart review. A Fisher’s exact test was used to compare radial artery graft usage between the prior comparable 6-month time period to the 6-month QI study period.

RESULTS

Quality improvement initiative results in increased utilization of more radial artery grafts used in isolated coronary artery bypass grafting cases

During the 6 months prior to the QI initiative, 252 isolated CABG operations were performed: 60 (23.8%) patients received a radial artery, 16 (6.3%) received a RIMA and 176 (69.8%) received a saphenous vein as a second graft. Therefore, a total of 76 (30.1%) isolated CABG operations were performed with multi-arterial grafting.

During the 6-month QI initiative, 247 isolated CABG cases were performed at our institution: 98 (39.7%) patients received a radial artery, 23 (9.3%) received a RIMA and 126 (51%) received a saphenous vein as a second graft. The total number of isolated CABG operations with multi-arterial grafting was 121 (48.9%). Therefore, radial artery graft usage significantly increased by 67% (P = 0.006) (Fig. 3) following implementation of the QI initiative.

Figure 3:

Incidence of radial artery as a second conduit in isolated coronary artery bypass grafting operations showing a significantly greater number of radial artery grafts used in the 6-month study period (P = 0.006).

The most common reason for not using a radial artery graft was inappropriate target vessel

The postoperative survey response rate was 93% (231/247). All 9 cardiac surgeons at our institution who perform CABG completed the survey at least once. The range of years in practice as an attending surgeon post-fellowship was 4–24 years (average ± standard deviation: 12.1 ± 7.1 years. Surgeons were permitted to select >1 reason for not using a radial artery graft which occurred in 17.4% of surveys (26/149).

The most common reason that a surgeon did not use a radial artery graft was poor quality target vessel or target vessel stenosis <80% (36.2% [54/149]) (Table 1). Other reasons were patient age >75 years (30.8% [46/149]), ejection fraction <35% (12.8% [19/149]) and renal insufficiency/dialysis (10.7% [16/149]). The least common reason was the lack of available staff to harvest the conduit (0.7% [1/149]); there were no technical challenges with radial artery harvesting (0% [0/149]).

Table 1:

Survey responses from cardiac surgeons for not using a radial artery conduit

Reason for not using radial artery grafta	Frequency of this reason
Age >75 years	30.8% (46/149)
Poor quality target vessel	27% (40/149)
EF <35%	12.8% (19/149)
Target stenosis <80%	10.7% (16/149)
Renal insufficiency/dialysis	10.7% (16/149)
Other (i.e. recent cocaine use, immunocompromised, phlebitis, patient request)	10.7% (16/149)
Relative contraindication (i.e. radial access for LHC, radial arterial line, arm trauma)	8.7% (13/149)
Unfavorable Allen test	4.7% (7/149)
Lack of availability of trained staff	0.7% (1/149)
Injury to radial artery during/after harvest	0% (0/149)

^aMultiple reasons could be selected by the survey taker.

EF: ejection fraction; LHC: left heart catheterization.

No significant complications from radial artery harvest

Due to the endoscopic radial artery harvesting training that occurred in the lead period prior to the QI portion of this project, all but 2 of the CABGs utilized endoscopic harvesting of the radial artery graft. Additionally, calcium channel blockers are frequently prescribed postoperatively to reduce radial artery spasm, so we examined its usage. Twenty-seven (28%) patients who received a radial artery graft were given a calcium channel blocker intraoperatively and 78 (80%) patients were prescribed a calcium channel blocker at discharge. There were no enduring or significant patient complications from radial artery graft harvesting, such as nerve injury, ipsilateral hand ischaemia, infection or bleeding recorded at outpatient follow-up (range: 6–601 days; median: 446 days).

DISCUSSION

This study demonstrates the utility of a QI initiative to successfully increase evidence-based usage of radial artery grafts in patients undergoing isolated CABG. QI interventions included (i) dissemination of updated evidence on radial artery outcomes, (ii) implementation of a radial artery decision-making algorithm and endoscopic radial artery harvesting training and (iii) establishment of a feedback mechanism to identify obstacles to radial artery usage via a postoperative survey. The QI initiative was conducted over a 6-month study period and resulted in an increase in the absolute usage of radial arteries as a second graft by 15.9%. Coronary artery anatomy (target quality and severity of stenosis) was the most common reason for not using a radial artery graft. No patients experienced complications from radial artery harvesting. We also identified operational barriers to use of the radial artery conduit.

Observational data, a large randomized controlled trial with long-term outcomes, and recent American College of Cardiology/American Heart Association revascularization guidelines support the use of multi-arterial grafting with the radial artery in CABG [16]. A review of the trends in radial artery usage in the STS database demonstrate an inflection point in 2018, although the use of arteries as second conduits remains the vast minority of cases compared to vein grafts [21]. A 2023 publication using the STS Adult Cardiac Surgery Database found that only 5.6% of national CABGs used bilateral internal thoracic artery and only 8.5% used radial artery grafts [18]. Moreover, only 14% and 21% of institutions were found to use >30 cases with bilateral internal thoracic artery or radial artery grafts respectively per study period [18]. The surgical community will require deliberate mobilization of resources to integrate radial artery usage into routine clinical practice. Here, we outline the steps of our QI initiative to help the surgical community implement the use radial artery grafts as a second conduit.

Reported barriers to adoption of radial artery use include the technical demand for radial artery harvesting and concern for ipsilateral hand ischaemia following harvest [15, 22]. Radial artery harvest requires detailed knowledge of forearm anatomy, which is lacking from training curriculums. Furthermore, endoscopic harvest requires specialized equipment and training. Nevertheless, endoscopic radial artery harvesting reduces harvest time, postoperative incisional pain and risk of wound infection [23, 24] and as such, we felt it was a critical aspect of our QI initiative. Finally, the true occurrence of ipsilateral hand ischaemia is rare due to radial and ulnar artery collaterals [25] and has been reported in only 4 patients over 4 studies analysing a total of 6719 radial artery graft harvests [26–29]. In our study, palmar arch patency was assessed with a preoperative and intraoperative Allen’s test [20], demonstrating zero incidence of hand ischaemia. Other groups have used Doppler ultrasound to assess for patency of the palmar arch and capacity to increase blood flow through the ulnar artery with occlusion of the radial [30]. However, no studies have reported a definite advantage of the Allen’s test or radial artery ultrasound over the other methodology.

Our QI initiative can be applied at other institutions to help implement evidence-based usage of radial artery grafts in CABG procedures. The use of radial artery grafts can improve CABG long-term outcomes in appropriately selected patients.

Limitations

Our study is a retrospective analysis from a single academic center over 6 months. Patients in the comparator groups before and after implementation of the QI initiative were not matched and multivariable analysis was not performed; therefore, we were unable to control for confounding variables. Moreover, while we encouraged algorithm usage as much as we could with giving updates in monthly section meetings, distributing the algorithm in clinics, incorporating it into the case posting questions asked by surgery schedulers and asking questions about usage in the postoperative survey, we ultimately cannot verify how closely the algorithm was followed by surgeons, PAs and surgery schedulers. Additionally, we reported the occurrence of intraoperative and postoperative usage of calcium channel blockers but did not examine the usage of other medications. Finally, biases such as observer bias and a Hawthorne effect cannot be ruled out in this study due to the methodology requiring surgeon buy-in for the radial artery usage algorithm and completion of the postoperative surveys.

CONCLUSIONS

Our QI initiative successfully increased evidence-based usage of radial artery as a second arterial graft and identified barriers to the usage of radial artery grafts. Radial artery QI initiatives can and should be implemented at other institutions to improve long-term outcomes after CABG in appropriately selected patients.

Glossary

ABBREVIATIONS

CABG

Coronary artery bypass grafting

IMA

Internal mammary artery

MI

Myocardial infarction

QI

Quality improvement

RIMA

Right internal mammary artery

STS

Society for Thoracic Surgeons

FUNDING

This work was not supported by any funding.

Conflict of interest: none declared.

DATA AVAILABILITY

Data are available on request.

Author contributions

Mary E. Moya-Mendez: Conceptualization; Data curation; Formal analysis; Methodology; Software; Writing—original draft. Isabel DeLaura: Data curation; Writing—review & editing. Steven W. Thornton: Data curation; Writing—review & editing. Adam R. Williams: Resources; Supervision; Writing—review & editing. Brittany A. Zwischenberger: Conceptualization; Methodology; Resources; Supervision; Writing—review & editing.

Reviewer information

Interdisciplinary CardioVascular and Thoracic Surgery thanks Ramón Aranda-Domene and the other anonymous reviewers for their contribution to the peer review process of this article.

 

Presented at the American College of Surgeons Clinical Congress Medical Student Program, San Diego, CA, USA, 16 October 2022.