Volume 134, Issue 1 , Pages 23-28, July 2007
Forty-one-month follow-up of the Symmetry aortic connector system for proximal venous anastomosis
Article Outline
Objective
Results of short- and midterm follow-up studies of the patency rate of the Symmetry aortic connector systems (St Jude Medical, Inc, Minneapolis, Minn) are controversial. Long-term follow-up studies are still lacking (so far, the longest mean follow-up period was 19 months). The aim of our study was (1) to evaluate the patency rate of this device over a longer time-period and (2) to analyze risk factors for graft occlusion.
Methods
Between November 2000 and July 2003, 76 Symmetry aortic connector systems were implanted in 42 patients. At follow-up, 24 patients with 44 mechanical connectors were studied with 64-slice cardiac computed tomography. Eight patients had died previously, 6 patients refused to undergo a computed tomographic scan, and 4 patients had to be excluded because of impaired renal function.
Results
From a total of 44 mechanical connectors studied, 24 (55%) were occluded, 20 (45%; confidence intervals 31%–61%) were patent, and 7 of these grafts showed stenosis in the area of the connector. Mean follow-up was 41 ± 10 months (18-52 months). Sex, age, left main stenosis, hyperlipidemia, hypertension, renal failure, target vessel, stenosis of the target vessel, diameter of the target vessel, type of surgical intervention, diabetes, ejection fraction, postoperative anticoagulation regimen, and the connector size showed no significant influence on the bypass graft patency (P > .05). The bypass graft flow was recognized to be the only risk factor for bypass graft occlusion (P = .0256).
Conclusion
Midterm follow-up data show a high number of occluded Symmetry aortic connector system vein grafts. On the basis of these observations, the use of the connector was abandoned at our institution.
Abbreviations and Acronyms: AVR, aortic valve replacement, CABG, coronary artery bypass grafting, MSCT, multislice computed tomography, SACS, Symmetry aortic connector system
Since CE Mark and Food and Drug Administration market approval were granted in the years 2000 and 2001, more than 40,000 Symmetry aortic connector systems (SACSs; St Jude Medical, Inc, Minneapolis, Minn) for proximal venous aortic connection have been implanted worldwide. Results of short- and midterm follow-up studies on the patency rate of these devices are controversial, with some reports showing good results and some an incidence of graft occlusion as high as 50% with the SACS.1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 Long-term follow-up studies are still lacking; so far, the longest mean follow-up period was reported by Kitamura and associates17 in October 2005. Twenty-seven SACSs were evaluated with multislice computed tomography (MSCT) by that group; the mean follow-up was 19 months and the 1-year patency rate of that mechanical device was 92.6%.
The goal of our study was (1) to investigate the patency rates of SACS grafts over a longer time period and (2) to analyze risk factors for graft occlusion.
Patients and Methods
From November 2000 to July 2003, a total of 42 patients at our institution received the SACS at the time of coronary artery bypass grafting (CABG) either alone (n = 18) or with aortic valve replacement (AVR) (n = 24). All proximal saphenous vein graft anastomoses to the ascending aorta were performed with the SACS (n = 76); conventional techniques were used for venous distal anastomoses and left internal thoracic artery grafts to the left coronary artery (n = 22). Indication for the use of the SACS was a localized calcification of the ascending aorta, which was identified by digital palpation of the aorta. Transesophageal echocardiography was done routinely. All operations were performed by the same surgeon, who had been certified after a St Jude Medical, Inc, course for the Symmetry device and had extensive experience with aortic valve, on-pump, and off-pump CABG surgery.
One of our patients had had a CABG operation 7 years earlier; none had undergone previous coronary stent implantation. A total of 4 patients underwent revascularization without cardiopulmonary bypass. Graft flow was always evaluated intraoperatively with a flowmeter (CardioMeds; Medi-Stim, Oslo, Norway) according to the transit-time principle. There were no device failures at the time of implantation.
Surgical Procedures
The saphenous vein was harvested by experienced surgeons using an open technique. Heparin was used to keep the activated clotting time in excess of 300 seconds. In off-pump CABG procedures, 150 to 200 IU/kg of heparin was given to reach a target activated clotting time of 300 seconds. In the on-pump cases and in combined AVR procedures, 300 IU/kg of heparin was administered. The veins were mounted on the correctly sized device by the trained surgeon depending on vein diameter. The proximal anastomoses were performed immediately after cardiopulmonary bypass was started, followed by crossclamping for the distal anastomoses or AVR. In on-pump CABG procedures, the proximal anastomoses were performed after heparin had been administered and the activated clotting time had reached 300 seconds.
Postoperative Anticoagulation
Intravenous heparin was always administered from postoperative days 1 through 5 to maintain the partial thromboplastin time at a value 2.5 to 3 times greater than the normal value. All patients undergoing CABG received long-term postoperative aspirin therapy (300 mg intravenously on the first postoperative day, then 100 mg/day orally). After biological AVR + CABG, patients received warfarin with an international normalized ratio target range of 2 to 2.5 for 3 months, followed by long-term aspirin treatment. Patients with mechanical AVR + CABG and those with permanent atrial fibrillation were put on a permanent regimen of anticoagulation.
Postoperative Follow-up and MSCT Evaluation
This study was approved by the institutional research ethics committee, and informed consent was obtained from all patients. In January 2005, all surviving patients who had received a SACS implant were contacted by telephone and invited to undergo a 64-slice MSCT (Siemens Somatom Sonata, Erlangen, Germany) to evaluate the bypass graft. Exclusion criteria were intolerance to contrast medium, creatinine clearance greater than 1.5 mg/dL, and claustrophobia. At that time, 8 of the 42 patients had already died, as will be described in detail below. Six patients refused to participate without giving reasons; all of them were under a cardiologist’s regular care and were in New York Heart Association class I. Four patients had to be excluded because of impaired renal function. The remaining 24 patients with 44 SACSs (ie, the study group) underwent 64-slice MSCT to determine patency of the bypass graft. Collimation was 0.6 mm and rotation time 0.33 seconds at 120 kV and 750 ms. The dimeric non-ionic contrast medium Visapaque 270 USB (Amersham Health, London, United Kingdom) was applied intravenously via an antecubital vein. The circulation time was measured with a 10-mL bolus injection before the main scan. The total contrast dose for the main scan was 80 mL. Each scan took 10 to 12 seconds; during this time the patients had to hold their breath, but this was never a problem. The complete examination took 10 to 15 minutes.
The raw data were reconstructed routinely in 1-mm slices showing 55% to 65% of the R-R intervals using retrospective electrocardiogram-gated reconstruction at 25 cm of the displayed field of view. The reconstructed images were transferred to a workstation (Wizard Siemens Medical Solution, Erlangen, Germany) for postprocessing. Axial, 3-dimensional volume-rendered, and maximum intensity projection reconstructed images were analyzed to evaluate the number, location, and patency of the bypass grafts. Graft occlusion was diagnosed if the graft was not visualized. The connector itself, however, was always visible.
Statistical Analysis
Numeric data were analyzed with the t test. Categorical data were analyzed by the Fisher exact test and χ2 analysis. Results are expressed as mean ± SD.
Results
Clinical Outcomes
Of the 42 patients, 3 died within the first 30 postoperative days; 2 of them, aged 80 and 81 years, underwent combined AVR and CABG and died of multiorgan failure in the intensive care unit. In both cases, autopsy showed all bypass grafts to be patent and valve function after biological AVR to be unimpaired. The third patient, aged 82 years, died 21 days after CABG of a pulmonary artery embolism at a rehabilitation center after he had been discharged from the hospital in good condition on the ninth posteroperative day.
Five of the remaining 39 patients died within the average observation period: 1 patient died of uterine carcinoma 20 months after surgery; 1 patient of a cerebrovascular accident 43 months postoperatively; 2 patients of myocardial infarction 3 and 11 months after surgery; and the fifth patient died of unknown cause 32 months after CABG. Nothing can be said about the patency of those bypass grafts. The study group comprised 24 patients (15 men and 9 women); their average age at the time of surgery was 70.9 ± 7.8 years. The clinical profile of the study group is summarized in Table 1. Among them, 44 SACSs and 10 left internal thoracic artery grafts were anastomosed. Twenty-one months after mechanical AVR + CABG with 3 mechanical connectors, 1 of these patients had to be reoperated on because all of the bypass grafts using a SACS were occluded. The redo procedure was performed off pump with 2 venous grafts to the left anterior descending and circumflex arteries. Proximal and distal anastomoses were performed with conventional hand suture.
TABLE 1. Clinical profile of the study group (n = 24 patients with 44 SACSs)
| Age (y) | |
 Mean ± SD | 70.9±7.8 |
| Range | 54–88 |
| Female (No. of patients) | 9 |
| Ejection fraction (%) | |
| Mean ± SD | 63.0±12.3 |
| Range | 35–80 |
| Left main trunk stenosis (No. of patients) | 5 |
| Three-vessel disease (No. of patients) | 6 |
| Two-vessel disease (No. of patients) | 3 |
| One-vessel disease (No. of patients) | 1 |
| Aortic stenosis + 3-vessel disease (No. of patients) | 5 |
| Aortic stenosis + 2-vessel disease (No. of patients) | 5 |
| Aortic stenosis + 1-vessel disease (No. of patients) | 4 |
| Coronary artery stenosis 75%-90% (No. of Cv) | 21 |
| Coronary artery stenosis > 90% (No. of Cv) | 23 |
| Old myocardial infarction (No. of patients) | 8 |
| Diabetes mellitus (No. of patients) | 5 |
| Hypertension (No. of patients) | 18 |
| Chronic renal failure (No. of patients) | 5 |
| Hyperlipidemia (No. of patients) | 8 |
Operative details of the study group are summarized in Table 2.
TABLE 2. Operative details for the study group (n = 24 patients with 44 SACSs)
| CABG (No. of patients) | |
| On-pump | 6 |
| Off-pump | 4 |
| CABG + AVR (No. of patients) | |
| Biological AVR | 11 |
| Mechanical AVR | 3 |
| Skin-to-skin time (min) | |
| Mean ± SD | 226±61 |
| Range | 82–371 |
| Bypass time (min) | |
| Mean ± SD | 124±48 |
| Range | 44–239 |
| Crossclamping time (min) | |
| Mean ± SD | 88±41 |
| Range | 15–193 |
| Left internal thoracic artery grafts (No.) | 10 |
| SVGs (No.) | 44 |
| Total distal anastomoses (No.) | 54 |
| Distal anastomoses per patient (No.) | 2.25±0.7 |
| Range | 1–3 |
| SACS size (mm) | |
| 4.5-5.0 (No.) | 19 |
| 5.0-5.5 (No.) | 23 |
| 5.5-6.0 (No.) | 2 |
| Intraoperative SVG flow (mL/min) | |
| Mean ± SD | 51±30 |
| Range | 17–152 |
| Target coronary arteries for SVG (No.) | |
| Left anterior descending | 11 |
| Right coronary artery | 16 |
| Circumflex artery | 15 |
| Diagonal branch | 2 |
| Coronary artery size (mm) | |
| <2 mm (No.) | 27 |
| >2 mm (No.) | 17 |
The mean follow-up was 41 ± 10 months (range 18–52 months). Twenty-four of 44 SACS grafts (55%) were occluded, 20 (45%; confidence interval 31%–61%) were patent. All of the hand-sewn left internal thoracic artery anastomoses on the left anterior descending artery were patent (n = 10). Seven (35%) of the open bypass grafts showed a stenosis between 50% and 70%; these were located in the area of the connector. Eighteen patients were in New York Heart Association class I, 4 in class II, and 2 patients were in class III. One of them had to be reoperated on because of an occlusion of all venous grafts with symptomatic unstable angina pectoris. All the other patients are under regular care of a cardiologist.
The perioperative details of the occluded and patent venous grafts are shown in Table 3. Intraoperative graft flow proved to be the only variable exerting a significant influence on graft patency (P = .0256).
TABLE 3. Data of patent versus occluded venous grafts
| Variable | Patent (n = 20) | Occluded (n = 24) | P value |
|---|---|---|---|
| Age (y, mean) | 71.7±6.9 | 70.9±7.9 | .7361 |
| Ejection fraction (%) | 60.3±16.2 | 58.5±15.6 | .7170 |
| Sex | |||
| Male | 9 | 15 | |
| Female | 11 | 9 | .3627 |
| Graft flow (mL/min, mean) | 61.3±34.8 | 41.2±22.5 | .0256 |
| Intraoperative pulsatility index | 2.69±1.19 | 2.50±1.05 | .6953 |
| Diabetes mellitus (No.) | 2 | 4 | .6731 |
| Left main trunk stenosis (No.) | 5 | 5 | .999 |
| Hyperlipidemia (No.) | 9 | 11 | .999 |
| Hypertension (No.) | 6 | 13 | .1350 |
| Value of coronary artery stenosis (%) | |||
| 75%-90% (No.) | 12 | 9 | |
| >90% (No.) | 8 | 15 | .2252 |
| Chronic renal failure (No.) | 0 | 5 | .0534 |
| Anticoagulation regimen | |||
| Long-term aspirin (No.) | 4 | 6 | |
| Warfarin, then long-term aspirin (No.) | 8 | 5 | |
| Long-term warfarin (No.) | 8 | 13 | .3800 |
| SACS size (mm, No.) | |||
| 4.5-5.0 | 9 | 10 | |
| 5.0-5.5 | 10 | 13 | |
| 5.5-6.0 | 1 | 1 | .9603 |
| Diameter of coronary artery (mm, mean) | 1.73±0.54 | 1.67±0.48 | .7093 |
| Target coronary arteries (No.) | |||
| Right | 10 | 6 | |
| Circumflex | 4 | 11 | |
| Left anterior descending | 6 | 5 | |
| Diagonal | 0 | 2 | .1094 |
| Surgical intervention (No.) | |||
| CABG | 8 | 9 | |
| AVR + CABG | 12 | 15 | .999 |
Discussion
Since the first clinical description of the SACS by Eckstein and associates18 in 2001, there have been highly varying reports in the literature on the patency rate for this mechanical connector device. There are reports of good results in short-term follow-up with patency rates of 96% on hospital discharge,3 89% patency on 3-month follow-up,5 and 87% patency on 7-month follow-up.16 This corresponds to the development in the first month of early postoperative graft thromboses with conventional suture technique that is reported to be 5% to 10%.19 In contrast, there are also reports of graft occlusion or a high-degree graft stenosis of 31% to 54% in short-term follow-up,6, 7, 8 wherein these reports covered only symptomatic patients, so that nothing can be said about the overall patency rate. A randomized study by Carrel and associates4 with a planned follow-up of 6 months was ended prematurely owing to a 38% incidence of stenosis in the proximal vein graft segment in those patients who received the SACS. Bergsland and colleagues9 showed an overall patency rate of only 50% after an average follow-up of 4.8 months. In contrast, 1-year patency rates of 80% to 86% were reported for conventional venous bypass grafts.19, 20
To our knowledge, the longest follow-up studies were those of Kitamura and coworkers17 (19-month follow-up with a 1-year patency rate of 92.6%) and Khan and associates21 (12-month patency rate of 84%). These studies, however, said nothing about the degree of stenosis of patent venous grafts. Long-term patency of venous bypass grafts stapled with the SACS has not as yet been reported in the literature.
In our average follow-up of 41 months, only 45% of the venous grafts created with the SACS were patent, and 35% of them showed stenosis of greater than 50% in the area of the device. All of the conventional left internal thoracic artery anastomoses on the left anterior descending artery were patent. Fitzgibbon and colleagues19 give the incidence of venous graft occlusions of conventional anastomoses after 1 year as 1% to 2.5% per year. Extrapolated for an average observation period of 41 months, this would be a patency rate of 71% to 92% for conventional grafts. Our results with venous grafts made with the SACS thus are poorer than would be expected with hand-sutured anastomoses.
Kinking of the graft has been mentioned as a possible reason for the high incidence of graft occlusion with the SACS.2, 3, 22 This is possible because the aortic connector is anastomosed at a 90° angle to the ascending aorta. Kinking can also be caused by an excessively long bypass graft inasmuch as, with mechanical anastomoses, unlike conventional ones, the proximal anastomosis must be created first, and so the length of the graft may be incorrectly estimated.23 In our study group, there were never any technical complications with the implantation of the SACS and all the proximal SACS anastomoses were patent at the end of the operation. We cannot evaluate whether graft kinking led to occlusion of the graft, because occluded grafts are recognizable in the MSCT as a stoppage of contrast medium flow and a kink in the bypass graft is not seen. We were, however, aware of the problem, and to avoid it, we placed the device laterally on the aorta.
Spontaneous restenosis of hand-sutured saphenous bypass grafts is described to occur at the distal part of the anastomosis. All of the stenoses in the open bypass grafts that were identified in our study were in the area of the connector, and there was no kinking of the vessels with these grafts. Therefore, a potential involvement of the device in the stenosis process is plausible.
The histopathologic study of Farhat and coworkers24 shows a further possible cause of graft occlusion. Connector application caused complete alteration of the endothelium in 58%; more than 80% of those showed severe lesions of the vascular wall, in 42% there was dissection of the media, and in 5%, a complete thrombus on the vascular wall.
As far as medical treatment to avoid early thrombotic event is concerned, there is no uniform regimen. Some use aspirin alone, others use aspirin and clopidogrel, and still others use an anticoagulation regimen for 3 months.8, 18, 25 The manufacturer’s recommendations for the SACS do not mention any antithrombotic treatment or anticoagulation after implantation. Our study did not show any significant difference in graft patency rate between antiplatelet or anticoagulant treatments, although it is uncertain whether clopidogrel would have influenced the occlusion rate.
Our study has some limitations. The patients in whom the connector device was used were not randomized; rather, they were chosen selectively on the basis of having partial arteriosclerosis of the ascending aorta. The reason for this was that this indication is mentioned in the literature and the use of the SACS was justified on the basis of good short-term follow-up results. Further, because of the high cost of the device, our division only had 80 SACSs at its disposal, which were applied selectively.
The expected patency rate for conventional hand-sutured proximal venous anastomoses was obtained from the scientific literature. Our historical controls of conventional anastomosis were similar to those results but were not published. Further, it remains unclear whether clopidogrel would have a positive influence on graft patency, as it does in interventional coronary stent application. None of our patients received this treatment and, to the best of our knowledge, there have been no relevant publications for SACSs up this to date.
Further, only 71% of the survivors underwent computed tomographic angiography, because 6 patients refused to participate and 4 patients had to be excluded owing to impaired renal function. These patients are under a cardiologist’s regular care, but nothing can be said about the patency rate of the grafts.
In conclusion, despite the promising results reported in the literature for short-term follow-up with patency rates similar to conventional suture technique, our poor results for patency rates in longer-term follow-up led us to abandon use of this device at our institution.
We gratefully acknowledge Peter Rehak, PhD, for statistical evaluation and Eugenia Lamont, BA, for language editing.
References
- . Severe stenosis of anastomoses by using the Symmetry Aortic Connector System. Ann Thorac Surg. 2004;78:1831–1833
- Video-assisted multivessel revascularization through a left anterior small thoracotomy approach with the Symmetry Aortic Connector System. J Thorac Cardiovasc Surg. 2003;125;:129–134
- Assessment of an aortosaphenous vein graft anastomotic device in coronary surgery (Clinical experience and early angiographic results). Ann Thorac Surg. 2002;74:2101–2105
- . Pitfalls and key lessons with the Symmetry Proximal Anastomotic Device in coronary artery bypass surgery. Ann Thorac Surg. 2003;75:1434–1436
- . Initial experience with sutureless proximal anastomoses performed with a mechanical connector leading to clampless off-pump coronary artery bypass surgery. Ann Thorac Surg. 2004;77:563–568
- Clinical, angiographic, and interventional follow-up of patients with aortic-saphenous vein graft connectors. Circulation. 2003;108:452–456
- Severe ostial saphenous vein graft disease leading to acute coronary syndromes following proximal aorto-saphenous anastomoses with the Symmetry bypass connector device (Is it a suture device or a “stent”?). J Am Coll Cardiol. 2004;43:133–139
- . Early bypass occlusion after deployment of nitinol connector devices. J Thorac Cardiovasc Surg. 2004;127:1421–1426
- Intraoperative and intermediate-term angiographic results of coronary artery bypass surgery with Symmetry proximal anastomotic device. J Thorac Cardiovasc Surg. 2004;128:718–723
- First-year outcomes of beating heart coronary artery bypass grafting using proximal mechanical connectors. Ann Thorac Surg. 2004;77:1542–1549
- First experience with the St Jude Medical, Inc, Symmetry Bypass System (Aortic Connector System). J Thorac Cardiovasc Surg. 2003;125:414–417
- . Early results of Aortic Connector System for proximal anastomoses of the saphenous vein graft. Ann Thorac Cardiovasc Surg. 2005;11:98–103
- Decreased intermediate term patency of automated proximal anastomoses evaluated by sequential ultrafast CT. Eur J Cardiothorac Surg. 2005;27:579–583
- . Angiographic results after the use of a sutureless aortic connector for proximal vein graft anastomoses. Ann Thorac Surg. 2002;73:193–194
- . Ultrafast computed tomography for quality control of automated proximal anastomoses. Heart Surg Forum. 2003;6:170–173
- Initial experience with proximal anastomoses performed with a mechanical connector. Ann Thorac Surg. 2003;75:1866–1871
- Early and midterm patency of the proximal anastomoses of saphenous vein grafts made with a Symmetry Aortic Connector System. J Thorac Cardiovasc Surg. 2005;130:1028–1031
- Minimizing aortic manipulation during OPCAB using the Symmetry Aortic Connector System for proximal vein graft anastomoses. Ann Thorac Surg. 2001;72:995–998
- . Coronary bypass graft fate and patient outcome: angiographic follow-up of 5065 grafts related to survival and reoperation in 1388 patients during 25 years. J Am Coll Cardiol. 1996;28:616–626
- . A randomized comparison of radial-artery and saphenous-vein coronary bypass grafts. N Engl J Med. 2004;35:2302–2309
- MDCT of the proximal anastomoses created by nitinol implants in coronary artery bypass grafting: a retrospective two-observer evaluation. Eur Radiol. 2005;15:305–311
- Port access video-assisted proximal anastomosis with the Symmetry Aortic Connector in MIDCABG procedure. Ann Thorac Surg. 2003;76:919–921
- . Flow dynamics of the St Jude Medical Symmetry aortic connector vein graft anastomosis do not contribute to the risk of acute thrombosis. J Thorac Cardiovasc Surg. 2004;128:117–123
- . Histological aspects of the saphenous vein damage with the use of the Symmetry Aortic Connector System. Interactive Cardiovasc Thorac Surg. 2004;3:373–375
- . Graft occlusion after deployment of the aortic connector Symmetry system: is anticaoagulation a necessity?. Ann Thorac Surg. 2004;78:1878–1888
Dr Bergmann
PII: S0022-5223(07)00361-3
doi:10.1016/j.jtcvs.2007.02.007
© 2007 The American Association for Thoracic Surgery. Published by Elsevier Inc. All rights reserved.
Volume 134, Issue 1 , Pages 23-28, July 2007
