Results of Endoscopic Radial Artery Harvesting in 1577 Patients

Dimitrova, Kamellia R. MD; Dincheva, Gabriela R. BS; Hoffman, Darryl M. MD; DeCastro, Helbert RN; Geller, Charles M. MD; Tranbaugh, Robert F. MD

Innovations: Technology & Techniques in Cardiothoracic & Vascular Surgery:
doi: 10.1097/IMI.0000000000000027
Original Articles

Objective: We reviewed 1577 consecutive patients undergoing coronary artery bypass grafting (CABG) using endoscopic harvesting of the radial artery (RA) to define our current results.

Methods: Since 2000, we have performed endoscopic RA harvest on 1577 consecutive patients; 1476 patients had isolated CABG, and 101 patients had CABG and other procedures. The mean ± SD age was 59.4 ± 9.0 years; 80.2% were men and 40% had diabetes mellitus. All data were prospectively collected. All-cause mortality was determined using the Social Security Death Index.

Results: There were nine in-hospital or 30-day deaths, for an operative mortality of 0.57%: mortality was 0.34% in isolated CABG and 3.85% in CABG/combined procedures. The overall estimated Kaplan-Meier survival at 1, 5, and 10 years was 99%, 95%, and 88%. In 37 patients, the RAs were not harvested or were not used for grafting because of a positive Allen test, extensive calcification or dissection, intramural hematoma, and scarring from previous arterial lines or catheterization. During postoperative follow-up, five patients (0.32%) were treated for incisional infection, and there were no ischemic hand complications. Three patients had a perioperative myocardial infarction in the RA graft distribution, and 15 patients had a coronary artery reintervention in the RA graft distribution. Two other patients had a percutaneous coronary intervention of their RAs. The overall RA patency at 10 years was 82%.

Conclusions: Endoscopic harvest of the RA is an excellent minimally invasive conduit harvesting technique with minimal morbidity.

Author Information

From the Division of Cardiothoracic Surgery, Beth Israel Medical Center, New York, NY USA.

Accepted for publication September 27, 2013.

Presented at the Annual Scientific Meeting of the International Society for Minimally Invasive Cardiothoracic Surgery, June 12–15, 2013, Prague, Czech Republic.

Disclosure: The authors declare no conflicts of interest.

Address correspondence and reprint requests to Kamellia R. Dimitrova, MD, Division of Cardiothoracic Surgery, Beth Israel Medical Center, 317 East 17th St, 11th Floor, New York, NY 10003 USA. E-mail:

Article Outline

Multiarterial grafting in coronary artery bypass surgery has been proven to remarkably improve the overall survival in these patients compared with saphenous vein (SV) grafting.1–3 Arterial grafts are associated with superior long-term angiographic patency compared with SV: in particular, the left internal thoracic artery (LITA) to the left anterior descending artery and radial artery (RA) to the left circumflex artery.4,5 However, despite the compelling evidence that RA is an excellent coronary graft and should be used more often in coronary artery bypass grafting (CABG), most CABG surgeries are still performed with SVs. The RA has been “avoided” because it is prone to spasm and its harvesting process is more difficult and lengthier compared with SV.

In a previous study, we found that both the endoscopic and open harvesting techniques yield RA conduits with excellent and equivalent midterm and long-term patency rates.6 Using the “no-touch” endoscopic harvesting technique7,8 with the application of an ultrasonic harvesting device instead of electrocautery9 and decreasing the ischemic time of the conduit immediately after the harvest10 protect the endothelial integrity of the vessel and prevent spasm. In the present study, we analyzed the long-term results of endoscopic RA harvest in 1577 consecutive patients.

Back to Top | Article Outline



An endoscopic vessel harvest program for RAs was started at Beth Israel Medical Center in January 2000. In the earlier years, the primary factors involved in the decision to use the RA conduit included the result of the Allen test, perceived patient life-span (patients <65 years), nonemergent nature of surgery, and a target vessel stenosis of at least 70%. However, in the recent years, the selection criteria extended to patients younger than 80 years and urgent and emergent operations. Unavailable or unsuitable SV conduits accounted for RA use in older patients. Most RA grafts were used to bypass the circumflex (73%) and diagonal (21%) arteries. Only 6% received RA to the right coronary artery. Renal failure was considered a contraindication for RA harvest because of concern for the need for possible upper limb dialysis access. The nondominant hand was usually chosen for the procedure except in those who had recently undergone invasive procedures involving the nondominant RA such as indwelling arterial line or transradial cardiac catheterization.

From January 2000 to October 2012, a total of 1577 patients underwent CABG with endoscopically harvested RA. Among them, 101 patients underwent CABG combined with an additional cardiac procedure (78 valve repairs or replacements, 4 transmyocardial revascularizations, 10 aortic root procedures, 3 left ventricle aneurysmectomies, 3 atrial septal defect repairs, and 3 left ventricular or biventricular assist devices), 40 underwent reoperative CABG, and 1436 underwent primary isolated CABG. The clinical and demographic data of these patients are shown in Table 1.

Radial artery use at Beth Israel Medical Center was very selective in the initial years, approximating 33% of all CABG surgeries, because the selection criteria were based on age younger than 65 years, elective operations, or unavailability of a venous conduit. Contraindications to RA use were hemodialysis or chronic renal insufficiency; Raynaud disease; and, recently, RA catheterization. Utilization gradually increased so that, currently, 75% of all patients receive an RA at BIMC using revised criteria of younger than 80 years and in urgent operations. The mean total RA use was 43% during the past 17 years.

All data were prospectively collected as part of the New York State Department of Health mandatory reporting system and maintained in a separate database at our institution. Institutional review board approval was obtained, and written informed consent was waived.

Back to Top | Article Outline
Surgical Technique

After induction of general anesthesia, an infusion of diltiazem (0.015 mg/kg per minute) was begun. The entire hand, forearm, and elbow were painted with alcohol beta-iodine or ChloraPrep (CareFusion, San Diego, CA USA) and draped. The wrist was hyperextended, with a towel roll underneath. A 2.5- to 3-cm longitudinal incision was made approximately 2 cm above the skin crease of the wrist, lateral to the palmaris longus tendon and slightly medial to the radial pulse. After exposure of the RA along with concomitant veins, extensive undermining dissection was carried out proximally to allow the subsequent insertion of the trocar. In the first few cases at the beginning in 2000, we used blind dissection with an Optical Vessel Dissector (VDC02) to create the tunnel and the anterior plane of the RA pedicle, but it resulted in bruising of the pedicle with a few torn side branches. The technique evolved from the use of the Optical Vessel Dissector to just using the Small Ultra Retractor (SVSR5) and stayed the same to date with a single distal incision and a medium 8 mm silicone round drain attached to a Bard closed wound suction evacuator. The Sorin endovein harvest system was used (Sorin Corp, Arvada, CO USA). A vessel loop was placed around the artery, and carbon dioxide was insufflated into the wound to fill up the perivascular sheath and to facilitate further dissection. The tunnel is extended using the same technique, enough to accommodate the hood of the SVSR5, maintaining a dry field. The SVSR5 is inserted, and the tunnel is extended to the antecubital area using 36-cm Harmonic Ace 5-mm laparascopic shears (HAR36). Once the tunnel is created, fasciotomy is extended to the antecubital fossa, carefully staying laterally to the pedicle and close to the medial border of the belly of the brachioradialis muscle. Then, the SVSR5 is pulled back, and the side branches, the lateral branches first followed by the medial, are divided with a harmonic scalpel and later controlled with surgical clips. If the medial branches are ligated first, the pedicle retracts laterally, making it difficult to ligate the lateral braches. After ligating the branches on both sides of the pedicle, the pedicle is rolled medially to expose the posterior branches using the SVSR5 and the HAR36. The posterior branches are ligated in the most careful fashion as a safe distance from the pedicle is kept with gentle retraction. The RA was dissected to its origin from the brachial artery first using endoloop Prolene, and the distal end is ligated using 0 silk tie. Incision is closed with 4-0 Monocryl.

Immediately after harvest, the RA conduit was placed in a solution of the patient’s blood (containing diltiazem), heparin (1 mg/mL), and papaverine solution (1 mg/mL) at room temperature; when papaverine is unavailable, only the patient’s blood with diltiazem and heparin (already in the circulation) is used. The same solution is used to distend the conduit by gentle intraluminal injection and clipping of the side branches. The RA conduit was stored in an identical solution until use. All patients received intravenous diltiazem or nitroglycerin (30–100 g/min) intraoperatively and for the first 24 hours. Distal anastomoses were performed first using continuous 7/0 polypropylene and the RA distended with blood. A single cross-clamp technique was used, and proximal anastomosis was usually direct to the aorta using a 4.5-mm punch and 5/0 or 6/0 polypropylene. The patients received aspirin (81 mg daily) starting on the first postoperative day and oral nitrates or calcium channel antagonist for 6 months. Predictably, the RA conduits are 15- to 25-cm long, easily reaching any vessel on the heart, and are an excellent size match to the native coronary arteries. Utilizing sequential or Y grafting increases the number of arterial grafts per patient.

Back to Top | Article Outline

The entire patient population was followed up to the first coronary angiogram for recurrent symptoms, date of death (using the US Social Security Death Index), or April 25, 2013. The primary endpoints of the study were all-cause mortality and perioperative major adverse events (MAE), which includes operative death, stroke, myocardial infarction, sternal wound infection, sepsis, reoperation for bleeding, respiratory failure, and renal failure. Operative mortality includes all patients who died in the hospital or by 30 days after operation; if discharged, MAEs were prospectively collected and were defined by the Department of Health Cardiac Surgery Reporting System ( All MAEs occurred during the index hospitalization, except for sternal wound and harvest site wound infection, which were reported up to 6 months postoperatively. A secondary endpoint was graft patency based on post-CABG symptom-driven angiographies.

Back to Top | Article Outline
Angiographic Analysis

Symptom-driven post-CABG angiograms were identified and analyzed by two observers independently. Conduits were evaluated on the first angiogram obtained after CABG and were classified as functioning if open and if the native vessel was fully opacified by the graft or as graft failure when there was a complete occlusion, a string sign (conduit 1 mm in diameter for some or all of its length), or a discrete stenosis exceeding 50% anywhere within the conduit or at either anastomosis.

Back to Top | Article Outline
Statistical Analysis

Demographic and angiographic data were collected prospectively and entered into computer database programs. In part, these databases satisfy the mandatory New York State Cardiac Surgery reporting system. Intraoperative data, including data related to the conduits, were entered at the time of operation. Postoperative angiographic patency data were entered at the time of angiographic review of the angiogram. Retrospective identification of the patients who underwent postoperative angiography allowed harvesting of the prospectively recorded data. Values are reported as mean ± SD. Percentages are given where appropriate. Statistical analysis was done with the XLSTAT software. A value of P < 0.05 was considered significant.

Back to Top | Article Outline


The demographic profile of the patients is presented in Table 1, and the MAEs are presented in Table 2. The hospital mortality was 0.57%. The mean number of grafts per patient was 3.9. The mean number of arterial grafts per patient was 2.5. Bilateral RA grafts were harvested in 6%; 8.2% of the patients received sequential and 9.1% received “Y” RA grafts. The resting length of the RA ranged from 15 to 25 cm.

In 37 patients, the RAs were not harvested or were not used for grafting because of a positive Allen test, extensive calcification or dissection, intramural hematoma, and scarring from previous arterial lines or catheterization. These patients were not included in the 1577 cohort analysis.

During postoperative follow-up, five patients (0.32%) had wrist wound infection, which required intravenous antibiotic treatment. Two patients required surgical debridement of the wound.

Grasping abilities and motor function were preserved in all patients. Mild sensory deficits and tingling sensation were common, but none had severe or disabling pain and/or marked numbness.

No patient sustained an injury to the ulnar artery or the brachial artery, and no patient had ischemia of the hand. Forearm mild ecchymosis was common after endovascular harvest of the RA but usually resolved 4 weeks after CABG.

Three patients had a perioperative myocardial infarction in the RA graft distribution, and 15 patients had a coronary artery reintervention in the RA graft distribution. Two other patients had a percutaneous coronary intervention of their RAs.

The estimated Kaplan-Meier survival of the entire cohort at 1, 5, and 10 years was 99%, 95%, and 88%, respectively. In the isolated primary CABG patients, survival at 1, 5, and 10 years was 99%, 95%, and 90%, respectively, which was significantly better than the survival in the CABG/combined procedures and redo CABG patients: 97%, 90%, and 78% respectively (see Fig. 1).

Of the total cohort of 1577 RA patients, 222 patients (14%) had symptom-driven cardiac catheterization at our institution 0.1 to 11 years after CABG, mean of years. The mean time to catheterization was 3.3 ± 2.7 years.

Table 3 shows the LITA, RA, and SV patency by territory grafted. Only 22 RA grafts were placed in the right coronary artery territory with an 82% patency.

Table 3 shows the angiographic patency per distal anastomosis for all conduits. A total of 1081 grafts (287 LITA, 15 RITA, 420 RA, and 364 SV) were evaluated in 278 patients (3.9 grafts per patient, 2.6 arterial grafts per patient). The overall LITA, RITA, and RA patency was 85%, 80%, and 82%. The RA occlusion rate was 16% (65 RA grafts), which included 7% with string signs (29 RA grafts) and 9% with complete occlusion (37 RA grafts) (Table 4).

Back to Top | Article Outline


The RA has been shown by many investigators to be a superb second arterial conduit in CABG, with excellent long-term patency compared with SV grafts.11–14 The better patency compared with SV has been attributed to a more attentive harvesting technique, the use of a pedicled conduit, a better selected target vessel quality, and indication of severely stenosed (>80%) coronary arteries. In addition, the RA conduit itself is relatively free from atherosclerosis and has metabolically active endothelial and smooth muscle layers, which contribute to the greater flow velocity and the ability to remodel accordingly to the targets’ size. In a previous study, we also found that the RA grafts are associated with a decreased disease progression in the native coronary arteries compared with the SVs, which likely contributes to the better survival in these patients.5

However, despite the excellent results with the RA, only 5% of CABG patients in the United States15 receive a second arterial graft and even fewer receive an RA graft. Its lack of “popularity” is because of the major concern about early vasospasm during and immediately after CABG. Elevated levels of a number of vasoconstrictors have been described, including endothelin-1 and angiotensin II.16 Interestingly, the proposed mechanism of the protective effect of the arterial grafts on disease progression and better overall patency has been attributed to the metabolically active endothelium, producing the same vasoactive and endothelial progenitor substances that defend the native vessels from progression of atherosclerosis17; in addition, incorrect handling of grafts during the manipulation may lead to vasospasm. The presence of a well-developed smooth muscle layer has also been blamed for the extreme vasoreactivity of the RA grafts, but worth of notice is that the excitation-contraction coupling in the smooth muscle cells is the main contribution to the conduit’s diameter adjustment to the coronary flow and the metabolic demand of the coronary circulation. Thus, being “alive,” metabolically active with its own vasa vasorum well-developed smooth muscle layer and endothelium, and a capability of “reacting” to the myocardial metabolic state and demand are most likely the RA attributes that make this conduit resistant to atherosclerosis and obliteration.

Reported short-term clinical outcomes of the endoscopic technique are excellent; however, because the endoscopic technique is concluded in a closed space, there are concerns raised about potential for mechanical or thermal injury to the endothelium of the arterial conduit. Shapira et al18 have assessed the effect of the endoscopic harvesting technique on the vasoreactivity and structural integrity of the RA by comparing it with the conventional harvesting techniques in a prospective randomized study. In that study, the authors demonstrated that structural integrity and the vasoreactivity of RAs harvested endoscopically remained intact, and also, they pointed out that the key points included harvesting the RA as a pedicle with the accompanying veins and fat, minimal manipulation of the graft (no-touch technique), and avoiding probing or hydrostatic dilation of the conduit.

Thus, the most important step for the successful application of the RA conduits is in the extremely careful handling during harvest. Adaptation of the endoscopic radial harvest no-touch technique and application of the ultrasonic harmonic shears minimize the microdamages to the RA. The use of ultrasonic harmonic coagulation and cutting shears established the mainstay of this less invasive, closed-space endoscopic procedure.19 No lateral spread of potential damaging energy to the RA pedicle or surrounding structures occurs. The blunt cone dissector is used to create a parallel tunnel surrounding the vascular bundle, and special care is always taken to avoid direct contact of the cone tip with the RA. During blunt dissection, the pressure of carbon dioxide is used to separate the tissue planes, and the RA procurement collects the vascular bundle as a whole. The concomitant veins help to prevent friction injury over the RA proper. Gentle flushing and soaking with a cocktail vasodilatation solution containing heparinized whole blood, calcium channel blocker, and papaverine help the conduit outside its natural environment. Postoperative use of oral calcium channel antagonist or nitrates in our patient cohort was also empiric for at least 6 months after CABG. All these measures resulted in 10-year survival of 90%. Only three patients (0.19%) had myocardial infarction in the RA graft distribution, and two patients (0.12%) had percutaneous coronary interventions done of the RA conduit.

Both open and endoscopic technique may cause injury to the lateral branch of the antebrachial cutaneous nerve or the superficial radial nerve; however, deficits of motor function were never encountered during postoperative follow-up in our cohort. Meticulous preoperative evaluation to ensure collateral circulation is a key to preventing this potential disaster. Mild sensory deficits were common in our study, usually presented with numbness in the territory of the superficial radial nerve in the early postoperative period. All of these patients experienced significant improvement after 3 months. Complete transaction of the superficial radial nerve did not occur in our series. The superficial wound infections were also very rare—encountered in only five patients (0.32%). The endoscopic RA harvesting technique was associated with very few wound infections, promoted earlier return to the daily routine, and improved patient satisfaction compared with open longitudinal RA harvest.

In conclusion, endovascular harvest of the RA is a technically feasible method that provides a superb arterial conduit for coronary revascularization, excellent cosmesis, and rapid return to normal activity. Providing a better conduit for coronary revascularization using this minimally invasive technique should always be considered. Radial artery use was associated with excellent long-term patency and patient survival.

Back to Top | Article Outline


1. Lytle BW, Blackstone EH, Sabik JF, Houghtaling P, Loop FD, Cosgrove DM. The effect of bilateral internal thoracic artery grafting on survival during 20 postoperative years. Ann Thorac Surg. 2004; 78: 2005–2012.
2. Tranbaugh RF, Dimitrova KR, Friedmann P, et al. Radial artery conduits improve long-term survival after coronary artery bypass grafting. Ann Thorac Surg. 2010; 90: 1165–1172.
3. Taggart DP, D’Amico R, Altman DG. Effect of arterial revascularisation on survival: a systematic review of studies comparing bilateral and single internal mammary arteries. Lancet. 2001; 358: 870–875.
4. Loop FD, Lytle BW, Cosgrove DM, et al. Influence of the internal-mammary-artery graft on 10-year survival and other cardiac events. N Engl J Med. 1986; 314: 1–6.
5. Dimitrova KR, Hoffman DM, Geller CM, Dincheva G, Ko W, Tranbaugh RF. Arterial grafts protect the native coronary vessels from atherosclerotic disease progression. Ann Thorac Surg. 2012; 94: 475–481.
6. Dimitrova KR, Hoffman DM, Geller CM, DeCastro H, Dienstag B, Tranbaugh RF. Endoscopic radial artery harvest produces equivalent and excellent midterm patency compared with open harvest. Innovations. 2010; 5: 265–269.
7. Souza DS, Johansson B, Bojö L, et al. Harvesting the saphenous vein with surrounding tissue for CABG provides long-term graft patency comparable to the left internal thoracic artery: results of a randomized longitudinal trial. J Thorac Cardiovasc Surg. 2006; 132: 373–378.
8. Souza D. A new no-touch preparation technique. Technical notes. Scand J Thorac Cardiovasc Surg. 1996; 30: 41–44.
9. Ronan JW, Perry LA, Barner HB, Sundt TM III. Radial artery harvest: comparison of ultrasonic dissection with standard technique. Ann Thorac Surg. 2000; 69: 113–114.
10. Emir M, Gol MK, Ozisik K, et al. Harvesting techniques affect the integrity of the radial artery: an electron microscopic evaluation. Ann Thorac Surg. 2004; 78: 1319–1325.
11. Zacharias A, Habib RH, Schwann TA, Riordan CJ, Durham SJ, Shah A. Improved survival with radial artery versus vein conduits in coronary bypass surgery with left internal thoracic artery to left anterior descending artery grafting. Circulation. 2004; 109: 1489–1496.
12. Desai ND, Cohen EA, Naylor CD, Fremes SE, for the Radial Artery Patency Study Investigators. A randomized comparison of radial-artery and saphenous-vein coronary bypass grafts. N Engl J Med. 2004; 351: 2302–2309.
13. Tatoulis J, Buxton BF, Fuller JA, et al. Long-term patency of 1108 radial arterial-coronary angiograms over 10 years. Ann Thorac Surg. 2009; 88: 23–29.
14. Buxton BF, Raman JS, Ruengsakulrach P, et al. Radial artery patency and clinical outcomes: five-year interim results of a randomized trial. J Thorac Cardiovasc Surg. 2003; 125: 1363–1371.
15. Tabata M, Grab JD, Khalpey Z, et al. Prevalence and variability of internal mammary artery graft use in contemporary multivessel coronary artery bypass graft surgery: analysis of the Society of Thoracic Surgeons National Cardiac Database. Circulation. 2009; 120: 935–940.
16. Borland JA, Chester AH, Rooker SJ, et al. Expression and function of angiotensin converting enzyme, chymase, and angiotensin II in the human radial artery and internal thoracic artery. Ann Thorac Surg. 2000; 70: 2054–2063.
17. Briguori C, Testa U, Riccioni R, et al. Correlations between progression of coronary artery disease and circulating endothelial progenitor cells. FASEB J. 2010; 24: 1981–1988.
18. Shapira OM, Eskenazi BR, Anter E, et al. Endoscopic versus conventional radial artery harvest for coronary artery bypass grafting: functional and histologic assessment of the conduit. J Thorac Cardiovasc Surg. 2006; 131: 388–394.
19. Brazio PS, Laird PC, Xu C. Harmonic scalpel versus electrocautery for harvest of radial artery conduits: reduced risk of spasm and intimal injury on optical coherence tomography. J Thorac Cardiovasc Surg. 2008; 136: 1302–1308.
Back to Top | Article Outline

This retrospective case series represents one of the largest reports of endoscopic radial artery harvesting in the literature. More than 1500 consecutive harvests were examined, with a mean ± SD follow-up of 6.9 ± 3.6 years. There were only five infections (0.3%) and no ischemic hand complications. The overall radial artery patency at 10 years was 82%. Late angiography was performed on a total of 420 radial arteries, with a mean ± SD time to catheterization of 3.3 ± 2.7 years.

This study suggests that in high-volume centers with a dedicated team, endoscopic radial artery harvesting can yield very good long-term patency rates with low complication rates.


Radial artery; Coronary artery bypass grafting; Endoscopic technique

©2013 by the International Society for Minimally Invasive Cardiothoracic Surgery