| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2000;17:407-414
© 2000 Elsevier Science NL
Cliniques Universitaires St-Luc, Avenue Hippocrate 10, B-1200 Brussels, Belgium
Corresponding author. Tel.: +31-71-526-3992; fax: +31-71-526-6965
e-mail: rdion{at}lumc.nl
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Patients and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
Key Words: Coronary surgery • Sequential grafting • Internal thoracic artery • Angiographic patency • Long-term follow-up
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
In 1989, we reported the clinical, functional and angiographic assessment 6 months postoperatively of the first 231 consecutive patients having received at least one sequential ITA graft from September 1985 until November 1987 at our institution [1]. Here, we have extended this initial cohort of patients to 500 in order to have sufficient a number of patients to be followed on the long-term from a clinical and angiographic point of view.
|
2. Patients and methods |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
1.5) and 12 (2.4%) with severe renal insufficiency (serum creatinine>2 mg/100 ml).
2.1. Patient selection
As previously described [1–3], nearly no situation existed that contraindicated the use of a single ITA graft aside from extreme chest deformities, severe disease of the subclavian arteries or of the ITA arteries or catastrophic emergencies. However, insulin-dependent diabetes, advanced physiological age, severe chronic obstructive pulmonary disease, obesity, non-viable myocardium or acute evolving myocardial infarction were regarded, in our institution, as either relative or absolute contraindications to the use of bilateral ITA grafting, depending on their mode of presentation or association. A patient less than 65 years of age normally benefited from complex arterial grafting whereas a patient more than 65 years of age had to somehow ‘deserve’ it.
2.2. Preoperative cardiac catheterization
Triple vessel disease was found in 281 patients (56%). The left ventricular ejection fraction (LVEF) estimated in a right anterior oblique (30°) view was less than 0.40 in 53 cases (11%). Routine preoperative angiographic assessment of both ITAs was required.
2.3. Operative technique
A 2-cm-wide pedicle of one or of two ITAs was harvested using low-power electro-cautery and metallic clips, starting at the level of the xyphoid cartilage and extending up to the subclavian vein. Further details of the techniques of ITA and saphenous vein harvesting and preservation, as well as the techniques of ITA and saphenous anastomoses have been described previously [1–3]. The ‘parachute’ technique with a running stitch of 8-0 polypropylene suture was used for all the ITA anastomoses which were fashioned as described previously [3]. In all, 1152 grafts corresponding to 2156 anastomoses (4.3/patient, maximum 7) were constructed. There were 1367 arterial anastomoses (2.7/patient) of whom 1350 were constructed with ITAs, 15 with gastroepiploic arteries and two with epigastric arteries. There were 1150 sequential ITA anastomoses (2.3/patient). Of the 320 RITA anastomoses, 269 (84%) were directed to the left coronary system: 71 (22%) to the LAD, and 198 (62%) to the Cx area. Our IMA grafting strategy has been repeatedly described previously [1–3]. We anastomose IMAs not so much to the best vessels, but to the best myocardium and we always try to keep the IMA graft pedicled. There were 243 bilateral mammary artery grafts.
Systemic hypothermia and intermittent cold crystalloid (St. Thomas) cardioplegia with topical cooling was used in all cases as described previously. The duration of cardiopulmonary by-pass averaged 116 min and aortic cross-clamp time 70 min.
2.4. Follow-up
Clinical follow-up was obtained by recalling the patients for an ambulatory examination or by telephonic contact and/or written questionnaire sent to patients, general practitioners and cardiologists.
One hundred and sixty-one patients consented to a late angiographic restudy after a mean postoperative interval of 7.5 years.
2.5. Statistical methods
Unpaired comparisons of proportions, Fisher-exact test was applied using Statxact. When needed, multiple proportions were evaluated by the Pearson chi-square test. No adjustment was made for multiple comparisons. The Kaplan–Meier curves were constructed using SPSS software.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
3.1.2. Early complications
Seventeen patients (3.4%) needed a marked inotropic support (
5 
dobutamine or dopamine) and two patients (0.4%) needed that inotropic support for more than 24 h.
A reoperation was necessary for a postoperative bleeding in 18 cases (3.6%), for persistent myocardial ischaemia in seven (1.4%), and for a sternal dehiscence in six (1.2%), two of them due to infection.
Perioperative myocardial infarction defined by enzymatic elevation (SGOT100) and/or electrocardiographic alteration was diagnosed in 18 patients (3.6%) with haemodynamic deterioration in two.
3.2. Follow-up
Follow-up was 97,4% complete (487 patients) and averaged 9.6 (range 8.6–13.6) years.
3.3. Late mortality
One hundred and twenty-five patients died during follow-up, 27 of a cardiac cause, 15 of a neoplasm, eight of vascular complications, and 16 of other miscellaneous causes. The exact cause of death could not be identified in 59 patients (mean age at follow-up was 71 years).
The Kaplan–Meier curve of the cumulated survival is displayed in Fig. 1. At 5 and 10 years, 89% and 72% of the patients, respectively, are still alive.
|
|
3.4.2. Reoperations
Nine patients (1.9%) needed additional surgery during follow-up: four coronary artery bypass grafts (CABG) (0.8%), two valvular procedures and three cardiac transplantations. Another 11 patients (2.3%) benefited from a percutaneous transluminal coronary angioplasty (PTCA), in six cases of a graft (five saphenous) and in five cases of a native coronary artery (four previously ungrafted). These iterative revascularizations are displayed in Table 1. Thus, in all, only 15 patients (3.1%) needed additional revascularization over the entire follow-up, an incidence of 0.3%/patient per year.
|
3.5. Angiographic results.
One hundred and sixty-one patients consented to a repeat catheterization at an average of 7.5 (range 1–12.2) years after the operation. All the angiographic restudies were reviewed separately by a cardiologist and a surgeon.
3.5.1. Overall angiographic patency rates of the arterial and of the saphenous anastomoses
These are illustrated in Table 2. Among the 448 arterial anastomoses, six GEA (intact), one IEA (intact) and 441 ITA anastomoses are included. A patent anastomosis is called ‘intact’ when no narrowing at all can be seen at angiographic restudy.
|
3.5.2. Comparison of LITA vs. RITA anastomoses, and of pedicled vs. free ITA grafts (Table 3)
There is no significant difference between the overall patency rates of pedicled LITA (97%) vs. RITA (93%) anastomoses (P=0.07, Fisher test), or between the sequential LITA/RITA anastomoses: 97 vs. 100% (P=0.4, Fisher test). When the ‘intactness’ is considered, there is indeed a significant difference between pedicled LITA and RITA (96.2 vs. 90.7%, P=0.05), but this only reflects the mediocre patency of the pedicled RITA to the distal RCA (15/18: 83%). There is a significant difference between the patency rate of pedicled ITA and free ITA anastomoses: 96.3 vs. 86.5% (P=0.02, Fisher test). The same is found when the intactness rates are considered.
|
The patency rates of LITA anastomoses are similar to those of RITA anastomoses for the LAD and Cx areas: 96.7 (264/273) vs. 96.3% (26/27) for the LAD, and 95.8 (46/48) vs. 93.2% (68/73) for the Cx, respectively. It is noteworthy that the long-term patency rates of RITA anastomosed to Cx (via the transverse sinus) are excellent (93.2%), as are those of RITA anastomosed to LAD (96.3%).
As mentioned previously [1], the patency rates of RITA to the RCA (usually distal) are less satisfactory (83%).
3.5.4. Sequential ITA anastomoses
These show excellent patency rates to all coronary vessels but the very distal Cx and the distal branches of RCA: 85% patency, 77% intactness (Table 4).
|
3.5.5. Patency of proximal vs. distal sequential ITA anastomoses (Table 5)
Overall there is no significant difference between the patency of the proximal and the distal ITA anastomoses: 96.2 vs. 96% (P=0.2, Fisher test). However, within the proximal ones, there is indeed a difference between the classical (L/L) and the diamond-shaped (#) anastomoses: 98 vs. 87.5% patency (P=0.02, Fisher test), 98 vs. 84% intactness (P<0.004, Fisher test), while this is not the case for the distal ones (P=0.1, Fisher test).
|
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
As found by others [6,12,13] the patency rates of RITA anastomoses to the distal RCA were less satisfactory (83.3%): they are identical to those found after 6 months [1]. Indeed, the 18 patients who were restudied at long term belong to this first cohort of patients. RGEA is currently (since 1988) preferred for grafting of the distal RCA with excellent results: 91% patency [14], as also reported by Grandjean et al. [15]. RITA is now restricted to the proximal RCA (down to the acute margin of the heart): by doing so, we were pleased to note that the patency rates of RITA to RCA returned to the level of those of RITA to LAD and Cx: 95.1% [2,3].
Sequential ITA anastomoses yield excellent patency rates to all coronary vessels except the very distal Cx and the distal branches of RCA (Table 4). This has also be found by others [4]. Therefore, for this area, we nowadays use either GEA or the free RITA in a ‘T’ connection with the pedicled LIMA [3,6,10,11,16]. In our experience (unpublished data), the early patency of the latter is identical to that of the pedicled ITA anastomoses to LAD and Cx.
Overall, free ITA grafts are significantly less patent than pedicled grafts as reported before [2,7], but this patency seems to be maintained over the long term, when compared to that obtained after 13 months [2] or after 15 months in our first 124 free ITA grafts [7].
Curiously, the inferior results seem to be exclusively related to the rather poor patency rates of the sequential free ITA anastomoses (80%) when the graft was connected with the ascending aorta (Table 3). As mentioned above, when a free ITA has to be used, it is nowadays reattached to the LITA, which, in our experience, seems to restore an excellent early patency. We believe that the advantage of this construction is twofold: not only the wall thickness is the same, but the free graft than is not exposed to the brute head of pressure emerging into the ascending aorta during the left ventricle systole.
In sequential ITA grafting, there is no significant difference in patency between the proximal and the last (distal) anastomoses (Table 5). When comparing the classical (constructed in the length) with the diamond-shaped anastomoses, there appears, in the long term, a difference in patency in favour of the classical ones, certainly when the latero-lateral anastomoses are considered.
Overall, the patency rates of the arterial anastomoses obtained after 7.5 years appear not much different from those obtained after 13 months [2] and after 6 months [1]: the attrition seems to be confined to the first months after the operation and is thus expected to be minimal for the following years.
Of course, native coronary disease is an evolving one: arterial grafts should therefore be evenly distributed over the myocardium and the use of sequential grafting should enable the surgeon to avoid reperfusing two adjoining vessels with two different arterial grafts. Also, the anastomoses should be placed distally to any coronary lesion of some significance, especially on the LAD, and, here again, sequential grafting must be considered in doubtful cases.
As mentioned previously [1–3], whenever the ITA flow insufficiency syndrome is suspected intraoperatively [16,17], instead of giving inotropes or of installing an intra-aortic balloon counterpulsation, we do not hesitate to add a saphenous vein graft in parallel with an ITA graft (usually to the LAD): 2% of all ITA grafts, with a markedly decreased frequency since we use warm surgery. This attitude has also been advocated by Carrel et al. [16]. Our strategies to address early postoperative ITA ‘insufficiency syndrome’ have been extensively described previously ([3,18,19: discussion R. Dion].
The evolution of our myocardial protection as from 1985 has been reported previously [3]. Since 1994, we converted to systemic normothermia and to intermittent antegrade warm-blood cardioplegia, as advocated by Calafiore et al. [20].
The conclusions of a prospective randomized study comparing cold crystalloid antegrade+retrograde and systemic hypothermia versus warm blood intermittent antegrade and systemic normothermia have been published recently [21] and have confirmed our new strategy. However, the extensive use of pedicled arterial grafts may interfere with myocardial protection, particularly in evolving ischaemia caused by the acute occlusion of a coronary vessel, and in a reoperation when an old pedicled arterial graft has remained patent. As reported before by us [3] and others [17], in these instances, either we add an intermittent retrograde warm-blood cardioplegia or we revert to the cold crystalloid antegrade and retrograde cardioplegia.
Eighty-four percent of the patients are still asymptomatic, and 71% are free of any type of ischaemia at 10 years. This compares favourably with results reported with bilateral ITA grafting by Pick et al. [22] and Fiore et al. [23]. Even more importantly, only 15 patients (3.1%) needed additional revascularization over the entire follow-up, an incidence of 0.3%/patient per year. When only repeat CABG is considered, this incidence falls to 0.08%/patient per year. Three out of the four repeat CABGs have been undertaken in patients having received a bilateral ITA at first operation (Table 1). One of them had even received a RITA to the LAD across the right ventricle anterior wall. Therefore, our indications for reoperation have been based on the severity of the recurrent angina and have not been influenced by any ‘reluctance to reintervene after arterial grafting’ as expressed by Sergeant et al. [24].
At 5 and 10 years, 92.8 and 69% of the patients remained free of any cardiac event (ischaemia, revascularization, transplantation and cardiac death). Again, this is comparable to the freedom of cardiac events reported after bilateral ITA grafting by Pick et al. [22] and Fiore et al. [23]: 46 and 66%, respectively, at 10 years.
These findings have encouraged us to increase our liberal use of ITA, and particularly of sequential ITA grafting, in the patients less than 70 years of age who are not crippled by severe comorbid disorders as mentioned above. Indeed, as stated by Sergeant et al. [25], patients in whom demographic, cardiac or non-cardiac comorbidity reduces the 10-year survival are unlikely to be improved by more than a single arterial graft, particularly to LAD.
The 500 patients of these series represented 23.5% of all the patients who have benefited from a CABG during the same time span. In 1998, the proportion of patients having benefited from sequential ITA grafting has climbed to 47%. Obviously, our confidence in applying ITA to the whole spectrum of grafting circumstances – using sequential, classical, diamond-shaped anastomoses, the transverse sinus route or ‘T’ grafts – has enabled us to enhance its versatility by deferring until the last moment the choice of its destination, according to intraoperative findings. The reward of this complex strategy has been, for our patients, an excellent clinical outcome and an extremely low need for repeat revascularization.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
Appendix A. Conference discussion |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
Dr Dion: On the operative protocol, we always have to mention whether the target coronary vessel is diffusely diseased or not, and I could have gone into it. The problem is that, with different surgeons, I am not sure that this estimation would have been discriminatory enough. But I agree that we probably should have looked at the preoperative angiograms to help in characterizing the distal disease.
Dr A. Moritz (Frankfurt, Germany): For the author, just a technical question. Did you ever use for your sequential work, which is sometimes expanded and you have to go behind the heart, a skeletonized mammary? I got the feeling that with the skeletonized graft, especially the sequential anastomosis, is made easier than with the pedicle. Did you ever try it or do you have any idea on this?
Dr Dion: Of course a skeletonized ITA is longer and is regarded by some surgeons as easier to handle. The point is that you also have more chances to get an intraluminal haematoma during harvest, and the handling during the anastomoses is also more delicate. Personally, over the last 6 months, I have converted to skeletonized thoracic arteries, especially in obese and diabetic patients, because of these reports about better sternal perfusion after harvesting. Therefore I can compare both methods, and I still believe that you ‘damage’ more arteries by taking them skeletonized than by taking them with a pedicle. We should then make a balance between that risk and that of a problematic healing of the sternotomy after bilateral harvesting in selected patients.
Mr J. Pepper (London, UK): What precautions do you take against conduit spasm and for how long do you give anti-spasm drugs when you use multiple arterial grafts?
Dr Dion: You mean after the operation?
Mr Pepper: Yes, after the operation.
Dr Dion: Well, spasm is difficult to ascertain after the operation. It is for me still a mystery. I believe that many of the so-called postoperative spasms of the ITA grafts have probably to do with too low a driving pressure for the coronary perfusion. After a multi-arterial revascularization, it is mandatory to keep a sufficient driving pressure. Personally I prefer a situation with 4 l of cardiac output and a mean systemic pressure of, let us say, 90 mmHg, than another one with 12 l of cardiac output and a mean systemic pressure of 45 mmHg. This is very important. Furthermore, we have to be very careful with the body oxygen consumption after extensive arterial revascularization. There are many reports showing that, for some time after the operation, the ITA graft sometimes lacks flow reserve after extensive revascularization. Therefore, we like to keep the patient sedated for at least 5 to 6 h after the operation and we don't like to extubate too soon, precisely not to exacerbate oxygen consumption. We have seen disturbing electrocardiographic changes disappear only by sedating the patient to diminish O2 consumption and/or by increasing filling pressures or stopping nitrates to increase coronary perfusion pressure. I believe that nitrates can be dangerous in some situations after extensive arterial grafting.
Dr J. Melo (Carnaxide, Portugal): Quite a number of your patients had lung disease. Could you comment on patient selection for sequential grafts and the tricks for tailoring a long enough graft to do it in COPD patients?
Dr Dion: In COPD patients we pay attention to preoperative FEV1. If FEV1 is below 1100 ml, we would not undertake a bilateral ITA grafting. We stick to a single ITA grafting which of course can be a sequential one. Indeed, in a COPD patient, one has to be careful with the length of the ITA because, when the lung is going to be reinflated, it will take up much room in the chest. Therefore, particularly in these patients, we incise vertically the pericardium and the pleura down to the phrenic nerve and the graft can then be routed just above the phrenic nerve and under the lung. Also, we free the ITA quite proximally in these cases in order to be able to divert it under the apex of the lung, which otherwise can be in the way when hyperinflated. Finally when building a sequential ITA graft in a COPD patient, we allow a somewhat longer ITA segment than usual between its branching off the subclavian artery and the first latero-lateral anastomosis.
|
References |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
