| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ann Thorac Surg 2002;73:701-703
© 2002 The Society of Thoracic Surgeons
a Département Cardio-Vasculaire, L’Institut Mutualiste Montsouris, Paris, France
* Address reprint requests to Dr Bachet, Département Cardio-Vasculaire, L’Institut Mutualiste Montsouris, 42 Blvd Jourdan, 75014 Paris, France
In this issue of the Annals of Thoracic Surgery, Westaby and colleagues [1] report an important series of operations for acute type A aortic dissection with an outstandingly low hospital mortality. In this report, the authors describe a steady, simple, reproducible surgical technique and conclude that hospital survival of patients operated on for acute type A dissection is mainly the result of the surgical technique and the surgeon’s skill. This article confirms a previous report published in 1997 in which the authors exposed similar results and drew the same conclusions [2].
The authors have to be congratulated for having obtained such impressive results in such a difficult surgical area and for having maintained those results throughout the years.
This article also confirms what the surgical community has attempted to demonstrate for years: surgical technique is of major importance in patient outcome. Thus the author’s answer to the question: "does the initial operation have a predominant influence on immediate and long-term results?," is a definite "yes."
To achieve this goal during the initial procedure, the aortic segment at risk (ascending aorta) the whole aortic root and aortic valve, the site and extension of the proximal intimal tear, and the consequent perfusion and dilatation of the false lumen all must be considered.
The proper management of the aortic root and aortic valve appears to be a major factor for satisfactory and stable results.
Several documented publications reporting large experiences have emphasized the necessity of preserving the native aortic valve whenever possible [3–5]. This is best performed by repairing the aortic root (possibly with the aid of glue) and resuspension of the valve [6, 7]. In addition to simplicity, this technique prevents the risk of a valvular prosthesis.
In some instances, though, the sinuses of Valsalva are either dilated or destroyed by the dissecting process. The only solution is to completely replace the aortic root. Failure to do so commits the patient to the risks of redissection or dilatation of the remaining sinuses of Valsalva, recurring aortic regurgitation, and late reoperation.
The simplest, shortest, and safest method is to replace the valve and the ascending aorta by means of a composite graft with reimplantation of the coronary ostia. It is obviously the technique of choice in patients with annuloaortic ectasia or Marfan’s syndrome [8].
But, in some instances, even though the sinuses of Valsalva are either dilated or destroyed by the dissecting process, replacing the valve might appear unjustified as it is normal or nearly normal.
In their article, Westaby and colleagues question the use of the techniques proposed by Sarsam and Yacoub [9] and David and Feindel [10] on the pretense that they are more technically demanding and time consuming and cannot be performed under emergency circumstances. Is it so? We have successfully performed these procedures in 10 cases, including three with Marfan’s syndrome. There was no hospital death, no particular bleeding and no neurologic complication. Residual aortic regurgitation was trivial in all patients except 1, who required late aortic valve replacement. These techniques may require further experience before being considered as routine procedures, but are very promising and may greatly benefit patients with complete disruption of the aortic root.
The second major element of operative success is the proper treatment of the intimal tear. An intimal tear located in the ascending aorta or transverse arch must be resected. Attempts to suture an intimal tear usually result in intraoperative death by hemorrhage or rupture or develop into an aneurysm of the ascending aorta and early reoperation. A few attempts have been made to glue the intimal tear without graft replacement [11, 12]. Immediate results were promising but there is concern that further dilatation of the nonreplaced aorta may occur and late reoperation may be required.
This leads to the important issue of transverse arch replacement during emergency repair. Because we consider resection of the intimal tear mandatory, it may be necessary to replace partly or totally the transverse arch [13–15]. In the great majority of patients, a partial or anterior "semiarch" replacement is sufficient, as the intimal tear is generally located in the concavity of the transverse arch. In our experience, as in that of the Oxford group (Dr Westaby’s group), emergency replacement of the transverse arch does not increase the surgical risk.
Does this justify systematic performance of an "open" distal aortic anastomosis, in cases in which the intimal tear is totally located in the ascending aorta? The possibility of inducing lesions (secondary intimal tears or rupture) with the aortic clamp has led many surgeons to avoid cross-clamping the vessel in any circumstance. They may be right, as the "open" distal anastomosis allows better exploration of the transverse arch and an easier and more extensive aortic repair. Nevertheless, this procedure necessitates deep hypothermia or cannulating innominate and left carotid arteries to perfuse the brain. By doing so, the patient is exposed to the risks of these techniques. But we now accept, after a long period of reluctance, the conclusion that, an open distal anastomosis, as advocated for years by Westaby and colleagues, is a key factor in improved results.
It has become apparent that the third important element of success may be intraoperative management of the false channel. Pressurization of the neo-lumen during cardiopulmonary bypass is often deleterious and may lead to intraoperative disruption of the distal aorta, severe malperfusion by compression of the true lumen, bleeding at the site of the distal anastomosis when cardiopulmonary bypass is resumed, etc. These complications are predominantly due to retrograde aortic perfusion through the femoral artery. This can be easily avoided by discarding femoral artery cannulation and by routinely using the right axillary artery. This provides physiologic antegrade perfusion of the aorta, throughout the whole procedure.
In view of our own experience and that of the most important papers published in the past decade, we agree with the Oxford group that: the immediate outcome of patients operated on for acute type A dissection "is dependent on the effectiveness of the surgical intervention" [2].
Nevertheless their article raises several questions.
To many surgeons, their results may appear "surrealistic". Most articles published during the last decade report hospital mortality rates of about 15% to 25% [6, 16, 17]. Perhaps more important is the fact that mortality rates have been rather stable for one or two decades. It seems that mortality rates have reached a lower threshold below which they cannot be reduced, despite many improvements in the management of these patients.
One of the reasons that may explain the difference in results reported by Westaby and colleagues and those of others may be the selection of patients. It is indeed ascertained that many factors in hospital death are totally beyond the capabilities of surgery and the surgeon’s control [18–20].
When acute dissection occurs, anatomic and physiologic damage are immediate, and within a few minutes or hours become irreversible and lethal.
Tamponnade is present in most cases. It does not always mean aortic rupture but always threatens the patient’s life and may be fatal if not relieved.
Malperfusion is much more insidious and dangerous, as the diagnosis is often delayed and treatment is difficult.
Preoperatively, malperfusion may be obvious (eg, neurologic disorder, limb ischemia). When diagnosed or suspected, the first treatment is, in our opinion, aortic replacement. In many cases, malperfusion disappears after aortic repair. However, it is often totally undiagnosed particularly when the digestive tract is concerned.
Intraoperatively, malperfusion often involves the brain and remains unknown during the entire surgical procedure in the absence of monitoring by electroencephalography, near infrared saturation, or transcranial Doppler.
Unrecognized or belatedly diagnosed malperfusions are responsible for many early deaths. In most cases, the lesions are irreversible when the diagnosis is made and most attempts to correct the ischemic damage are in vain.
In some patients operated on rapidly after the onset of acute dissection, and in particular within the first 12 hours, the insidious process of visceral malperfusion may already be initiated but unknown, as neither clinical symptoms nor biological signs are present. Despite a perfectly executed, straightforward surgical intervention, the unrecognized or belatedly recognized irreversible lesions will be fatal. If, on the other hand, the patient is operated on the second or third day after the onset of symptoms, severe tamponnade, stroke, and major visceral malperfusion are excluded. Thus, because of natural selection, the patient may be among those surviving the 50% mortality rate in the first 48 hours. As acknowledged by Westaby and colleagues in their article, this could be the case in their experience, since 40% of patients were operated on more than 24 hours after the onset of symptoms.
Another selection bias may be that all patients included in this series were operated on by the principal author. Generally articles report the whole experience of a given surgical group, which includes all surgeons of various ages and experience, a common surgical policy, a certain technical homogeneity, and more or less comparable results. It is possible that the results obtained by the most experienced surgeon are better than those of the whole group, but this rarely appears in the statistics. One may presume that the senior surgeon is not on call every night and every weekend of the year and, therefore, that many acute type A dissections are operated on by other surgeons. It might even be possible that the most urgent, moribund patients are operated on in the middle of the night by younger colleagues on duty, and that some patients in good clinical condition who can wait are operated on when the senior surgeon is available. If the overall mortality rate of the whole Oxford experience is similar to that of Mr Westaby, then it surely represents a major breakthrough in the surgery of acute type A dissection. If they are closer to those found in the literature, it would be comforting for the rest of the surgical community and, more importantly, a better reflection of the surgical risk for patients experiencing acute type A dissection in the Oxford area.
Compared with the average rate of about 15% to 20% reoperations at 10 years [16, 21], the very low rate of late reoperation in Mr Westaby’s experience requires more information. There is no doubt that the quality of the initial emergency procedure has a clear influence on late results. One can therefore state that a more appropriate surgical repair of the proximal aorta may reduce the rate of late reoperation and distal rupture. Those two complications are the cause of up to 50% of late deaths [18].
Unfortunately, the improved prognosis produced by better surgical management may not apply to patients with Marfan’s syndrome. The necessity for late reoperation or the occurence of late rupture in these patients is primarily linked to the basic disease and to pathologic progression in the distal aorta. It is unlikely that this progression is greatly affected by better management of the proximal vessel during the first emergency procedure except, perhaps, by reducing false lumen perfusion by using antegrade flow. Whatever technique is used during the first emergency operation, Marfan patients are at a high risk for late reoperation. In our experience, Marfan patients, who represent 13% of all patients operated on for acute type A dissection, have a relative risk of late reoperation sixfold higher than that of non-Marfan patients (0.72 versus 0.12), despite the fact that they all had complete replacement of the aortic root with the Bentall or Yacoub technique [6, 18].
Is it possible to reduce the prevalence of late reoperation or distal rupture in Marfan patients? We know that they must be closely followed and that they must be maintained on ß-blocking therapy. The best management, however, would be "prophylactic" operation before they experience acute dissection. Too many patients are not operated on in time for a diagnosed aneurysm of the ascending aorta, even though operation is advised as soon as dilatation reaches 45 mm to 50 mm in diameter or shows signs of progression.
A very difficult question remains unsolved: when should we reoperate on patients with a residual thoraco-abdominal dissection, particularly if they are totally asymptomatic and live a normal life? The delicate balance between the risk of rupture and the risk of thoraco-abdominal aortic replacement still represents a difficult and troublesome dilemma.
Presently, nobody denies that acute dissections involving the ascending aorta require urgent operation to allow survival of the acute episode. It is also obvious that, during the past two decades, simple rescue of the threatened patient has been extended to the more ambitious purpose of removing diseased aorta as completely as sensible and preventing late complications. We agree that the initial surgical technique is of paramount importance in this matter. We have reached the conclusion that these goals are best achieved by adapting surgical techniques to pathologic features. Yet, outstandingly good results obtained by an experienced surgeon may not represent the real surgical risk. They can be misleading to some colleagues or can be used ill-advisedly by inexpert observers (eg, lawyers or health public service supervisors).
It appears that, in many centers, late mortality and morbidity rates have not been dramatically reduced during the last two decades. This may be related to the inadequacy of the surgical techniques used. It is more likely due to immediate anastomotic and physiologic damage, belated diagnosis of complications, and in many instances to the extremis condition and basic disease of the patient. Unfortunately these elements are not under the surgeon’s control. We wish they were!
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 |
