Abstract
Objectives: The severely calcified so-called porcelain aorta is one of the most dangerous and challenging findings in patients requiring coronary bypass surgery. Several techniques and technologies have been invented to handle this potentially lethal disease. We report on our initial experience with the PAS-Port® automated proximal clampless anastomotic device (Cardica, Inc., Redwood City, CA, USA), especially focussing on these patients. Methods: PAS-Port® anastomoses (for saphenous vein grafts) were performed in 17 patients undergoing coronary artery bypass graft (CABG) surgery. Of these, eight presented with the entity of porcelain aorta. In two patients, the diagnosis was previously known, in six cases heavily calcified aortas prohibiting any clamp manoeuvre were incidentally found intra-operatively. The site of anastomosis was determined by palpation and in individual cases with epi-aortic echocardiography. Other indications for PAS-Port® were localised dissection, acute myocardial infarction and partial sternotomy. Multislice computed tomography (CT) was performed in every patient to evaluate graft and anastomoses patency and appearance. Results: All 25 PAS-Port® anastomoses were triggered successfully. Two patients developed neurological deficits (prolonged reversible ischaemic neurological deficits, (PRIND)), with use of cardiopulmonary bypass (CPB) being the major predisposing factor (p = 0.02). Graft patency could be affirmed in all grafts by multislice CT in all patients. Conclusions: PAS-Port® anastomoses can be performed quickly, easily and, above all, safely in conditions prohibiting aortic clamping. Short-term results are excellent. Clear visualisation of anastomoses using multislice CT is an important advantage of the PAS-Port® device.
1 Introduction
One of the most feared settings in coronary artery bypass surgery, the heavily calcified so-called porcelain aorta, can be overcome by the combination of innovative techniques such as total arterial revascularisation, T-graft anastomoses and off-pump surgery [1–3]. Although still demanding, widespread use of these (aortic) ‘no-touch-techniques’ has reduced the incidence of neurological complications in cardiac surgery [1,4–8]. Alternatively, hybrid procedures also aiming to avoid any aortic manipulation may be useful.
However, daily practice demonstrates that a porcelain aorta may be found intra-operatively by chance, coming upon an unprepared surgeon as well as an unprepared patient. At that time of surgery, usually the internal thoracic artery as well as the saphenous vein are harvested and both arms are neither sterile nor have the radial arteries been examined sufficiently preoperatively.
In this study, we evaluated the use of the PAS-Port® automated aortic connector especially in the context of the porcelain aorta.
2 Material and methods
We started using the PAS-Port® device when planning surgery for a high-risk patient with a heavily calcified ascending aorta needing complete revascularisation for severe three-vessel coronary artery disease and contraindications for total arterial grafting.
2.1 Patients
A total of 17 patients received PAS-Port® anastomoses between August 2007 and January 2009. Patient characteristics are demonstrated in Table 1 . Mean age was 72 ± 9 years; 10 patients were male.
Patient characteristics.
In total, eight patients had aortic calcifications, either circumferential or segmental (Figs. 1 and 2 ), prohibiting any clamp manoeuvre. Six of these were not diagnosed until intra-operative palpation. One patient underwent scheduled off-pump coronary artery bypass (OPCAB) surgery prior to trans-apical aortic valve implantation.
Extent of shell-like aortic calcification in chest X-ray (a) and angiography (b). Visualisation of PAS-Port® anastomoses and aortic calcification in multislice CT (c–f) (a–e all same patient).
(a–c) Extent of aortic calcification in different patients. (d–e) Location of PAS-Port® anastomoses (black arrow) close to calcification, Y-graft, LITA (*).
The PAS-Port® device proved to be useful in another nine patients without porcelain aorta. For OPCAB surgery, one has the benefit of getting a fast proximal anastomosis with immediate inflow. In one patient, OPCAB was performed with partial lower sternotomy because of permanent tracheostomy after larynx surgery, to reduce the risk of sternal infection. The aorta could hardly be clamped, but was easily accessible with the PAS-Port® device. In another patient with constrictive pericarditis years after coronary artery bypass graft (CABG) surgery, a saphenous vein graft was accidentally injured. To avoid meticulous preparation of the ascending aorta for side-clamping, a new short vein segment was connected to the ascending aorta using PAS-Port®, with the distal end anastomosed to the blunt end of the injured graft. Another patient underwent emergent coronary bypass surgery because of acute dissection of the proximal right coronary artery with distinct retrograde expansion into the right coronary cusp. Side-clamping might have triggered aortic dissection, so the proximal anastomoses were performed using PAS-Port®. Immediate graft perfusion enabled fast target vessel supply.
2.2 PAS-Port® device
The PAS-Port® automated proximal anastomotic connector can be seen in Fig. 3 . It is a CE-marked single-use device for automated proximal vein graft anastomoses [7,8]. The saphenous vein can be harvested as usual but without the use of ligaclips, which may hitch within the device and disrupt the vein. Sealing of side branches can be performed either with ligature or, after deployment, with ligaclips. The design requires the proximal anastomosis to be performed first. The calibre of the vein should be 4–6 mm outside diameter; a test bench is provided with the device. The vein is pulled through the device and the proximal end is everted and pierced through the deployment clasp. After determining the position of the proximal anastomosis, the device is placed on the aorta and deployment is triggered by simply turning the release button. The width of the PAS-Port® anastomosis is 10 mm, and a suitable area of non-calcified aortic wall of 15–17 mm (‘finger tip’) in diameter suffices for deployment. The anastomosis is generated within 1–2 seconds and the graft perfused immediately. The graft originates at 90° to allow for variable take off angles, which is additionally facilitated by the fact that the anastomosis is performed with an everted vein end so that the pivot point is 1–2 mm below the external aortic wall (Fig. 3). One major advantage of the PAS-Port® system is that no foreign material reaches into the vein lumen (Fig. 3).
PAS-Port® connector for automated (proximal) vein-to-aorta anastomoses.
It has to be noted that the device should not be used without detailed theoretical and practical training, especially for intra-operative control of anastomosis bleeding.
2.3 Computed tomography (CT) scan
All patients underwent postoperative control using multislice (64-slice) CT within 30 days after surgery to determine graft patency.
2.4 Statistical analysis
Statistical computations were carried out with the commercially available software SPSS 17 using Student’s t-test for unpaired samples and Fisher’s exact test.
Continuous variables were expressed as mean ± SD, and qualitative variables as proportions (percentages). Effects were assessed as significant, if the explorative p-value was below the 5% margin.
3 Results
The design of this study is retrospective. All PAS-Port® anastomoses were performed by the same surgeon.
Patient characteristics are summarised in Table 1.
Patients with porcelain aorta were characterised by older age (75 ± 3 years vs 68 ± 12, p = 0.13, n.s.) and higher mean logistic EuroSCORE (27 vs 9, p = 0.1, n.s.). The mean number of anastomoses was 1.5 per patient.
Patients with porcelain aorta either underwent preoperative CT-scan or epi-aortic ultrasound (Fig. 4 ) additionally to manual palpation to identify aortic areas at presumably low risk for embolic events for anastomoses deployment. Trans-oesophageal echocardiogram (TEE) of the descending aorta also gave an indication of the extent of the disease (Fig. 4). In one patient, a Y-graft was used o minimise aortic manipulation (Fig. 2(f)). All patients had an in situ left internal thoracic artery (LITA) graft to the left anterior descending (LAD).
TEE of descending aorta with mobile plaque (a), epi-aortic echocardiography showing dorsal (b) and lateral calcification (c).
Aortic cannulation for cardiopulmonary bypass (CPB) was necessary in two patients with a severely calcified aorta. One of them was thought not to tolerate any manipulation of the heart otherwise; the other (emergent surgery for acute myocardial infarction) had to be converted from initially favoured OPCAB to on-pump beating heart CABG because of refractory ventricular fibrillation. Sites of aortic cannulation were determined by manual palpation.
One patient suffered from postoperative delirium for several days. Adverse cerebrovascular events occurred in two out of the 19 patients. Both demonstrated complete recovery, as proven by clinical and cranial CT-scan evaluation. Events were classified as prolonged reversible ischaemic neurological deficits (PRIND). Both patients had a severely calcified aorta and underwent surgery with the use of CPB.
Six patients with porcelain aorta were extubated within the first day postoperatively, one at day 9, while one patient needed tracheotomy. Mean intensive care unit (ICU) stay was 10.5 ± 12.0 days, mean hospital stay 22.4 ± 9.5 days. There was one in-hospital death (day 33, non-cardiac).
All other nine patients were extubated within the first day after surgery. Their mean ICU stay was 1.8 ± 2.0 days, mean hospital stay 11.3 ± 4.9 days. One patient died at day 7 from non-cardiac complications.
None of the patients without porcelain aorta demonstrated any neurological anomaly.
In this study, the concomitance of porcelain aorta and use of CPB was the major predisposing factor for adverse neurologic events and prolonged ventilation time (p = 0.02). Differences in ICU stay between patients with or without porcelain aorta were also statistically significant (p = 0.04, hospital stay p = 0.08, n.s.).
All PAS-Port® anastomoses were triggered within 1–2 seconds and could be deployed successfully. One anastomosis had to be ligated because the vein was oriented upside down. Bleeding problems, if any, could be easily controlled by a simple aortic purse string suture (5/0 prolene) around the graft origin. Early patency could be affirmed in all grafts by multislice CT (Fig. 2).
4 Discussion
The severely calcified or ‘porcelain’ aorta may present with either circumferential or mostly cloddy calcification. Both entities are greatly feared in cardiac surgery because of their often lethal complications resulting from embolisation. It is crucial to reduce any aortic manipulation to a minimum, that is, to avoid cannulation for CPB as well as aortic cross- or side-clamping [2,3]. Aortic no-touch techniques, consisting of OPCAB surgery with total arterial revascularisation, including T-graft anastomoses, have proven to reduce the risk of embolic events [1,9]. However, daily practice demonstrates that neither are all patients with porcelain aorta selected preoperatively, nor are those diagnosed intra-operatively by palpation prepared adequately for these techniques. Urgent or emergent cardiac operations, prohibiting to some degree meticulous patient preparation, may further increase the accidental finding.
Several veno-aortic anastomotic devices, either automated or manual, have been developed for CABG surgery, aiming to achieve homogenous anastomoses, to facilitate minimal invasive procedures and to avoid the need for aortic clamping [4,7,10].
Mostly technical complexity for manual devices and graft failure for automated connectors, not mentioning the costs, have resulted in their limited use. The St Jude Symmetry device, for example, had to be withdrawn from the market in 2004 because of early graft stenosis and a high incidence of major adverse cardiovascular events, attributed to the foreign body within the small vein lumen [6,8,11,12]. Nonetheless, there is still a need for such tools, especially when dealing with aortic calcification.
In this study, we evaluated the use of the PAS-Port® automated and clampless proximal anastomotic device, especially focussing on patients with severe aortic calcification. Safety and efficacy of the device have already been reported by others, with an equal freedom from major adverse cardiac events (9-month follow-up) of 97% for PAS-Port® and hand-sewn anastomoses, a short-term (discharge, single-centre experience on 51 patients) patency rate of 100% and intermediate term (9 month, multicentre experience on 183 patients) patency rate of around 80%, being non-inferior to hand-sewn (venous) anastomoses [7,8,13]. This is attributed to the unique design of the connector allowing the lumen of the vein to be free from any foreign material [7]. Short-term results of the PAS-Port® anastomoses in this study were also excellent. Patency of all grafts could be demonstrated with multislice CT within 30 days after surgery, with clear visualisation of the anastomoses.
We could demonstrate that the PAS-Port® device can be used safely even in a severely calcified aorta, following meticulous determination of graft origin by manual palpation, at best combined with careful epi-aortic ultrasound. The latter can easily be conducted with a 10–15 MHz linear array vascular transducer, allowing for good near-field visualisation. Hereby also, soft plaques may be detected. In our series, all PAS-Port® anastomoses could be deployed successfully. Reversible neurological deficits occurred in two patients, the only ones with porcelain aorta and use of CPB. All other patients experienced no neurological event. This is similar to the study by Kapetanakis et al. [9], comparing different practices of aortic manipulation, and to the study by Kempfert et al. [7], reporting on 51 patients with PAS-Port® anastomoses without any neurological events.
Aortic cannulation for CPB must be avoided by all means in these patients. Most likely it is not only the – cautiously determined – cannulation site per se, but also altered flow conditions including some kind of sandblast-effect on aortic plaques, leading to catastrophic embolic events [9].
While working with the PAS-Port® device, several other useful options arose. For example, at redo cardiac surgery, aortic preparation can be reduced to a minimum. Furthermore, in minimal access surgery with lower sternotomy aortic side-clamping may be difficult, and proximal CABG anastomoses can be better performed with the straight PAS-Port® device, even if bulky. Finally, subsequent trans-apical aortic valve implantation is certainly an option for patients with combined coronary and aortic valve disease.
5 Conclusion
PAS-Port® anastomoses can be performed quickly, easily and, above all, safely in most conditions prohibiting aortic clamping. Short-term results are excellent. Clear visualisation of anastomoses using multislice CT is an important advantage of the PAS-Port® device.
6 Limitation
The sample size of this study is small. This is attributed to the fact that most patients with porcelain aorta were treated with aortic no-touch techniques (OPCAB with T-graft anastomoses). A randomised trial comparing different modes of aortic manipulation or PAS-Port® anastomoses in patients with heavily aortic calcification would be unethical.
