Eur J Cardiothorac Surg 1999;15:32-38
© 1999 Elsevier Science NL
Minimal access aortic valve surgery 1
Christian L Olin*,
Árpád Péterffy
Department of Cardiothoracic Surgery, Linköping Heart Center, University Hospital, S-581 85 Linköping, Sweden
* Corresponding author. Tel.: +46-13-224-800; fax: +46-13-100-246; e-mail: christian.olin@thx.us.lio.se
 |
Abstract
|
|---|
Objective: We report our experience with minimal access aortic valve surgery and discuss the three approaches used. Methods: From June 1996 to October 1997, 18 patients underwent minimally invasive aortic valve surgery through three different incisions: right parasternal minithoracotomy (three cases), upper ministernotomy (11 cases), and transverse sternotomy (four cases). No special surgical instrumentation was used. Aortic valve replacement was carried out in 17 patients and aortic valve repair in one patient. The patients ranged in age from 42 to 86 years (mean 64 years). Concomitant procedures involving the aortic root and the ascending aorta were performed in five patients. Results: There was no mortality and no complications related to the procedure or the access. There was no instability or paradoxical movement of the chest wall. One patient was reoperated for postoperative bleeding. All patients were discharged from hospital within the usual time. No attempts were made to discharge them earlier, even if they recovered quickly. Conclusions: Of the three incisions used, the upper ministernotomy seemed to be the safest and easiest to perform. Through this incision, both the aorta and the right atrium could be cannulated, the right ventricle was accessible, and concomitant procedures on the ascending aorta could be carried out. The drawback of minimal access aortic valve surgery in general is that it is difficult to de-air the heart and more difficult to master intra- and postoperative complications should they occur.
Key Words: Cardiac surgery • Aortic valve surgery • Minimally invasive cardiac surgery
|
1. Introduction
|
|---|
Minimal access aortic valve surgery theoretically has many potential advantages compared to valve surgery performed through conventional median sternotomy. These advantages include less surgical trauma, less bleeding, decreased pain, reduced risk of wound infection, shorter hospital stay and faster rehabilitation. The small incisions are also cosmetically attractive to patients. The avoidance of manipulation of the heart results in fewer pericardial adhesions and facilitates reoperation. However, there are also many disadvantages with limited access. The approach can only be used in selected cases and demands considerable surgical experience, especially if intraoperative complications occur. The purpose of this paper is to present our experience with three different incisions and discuss their advantages and disadvantages.
|
2. Materials and methods
|
|---|
2.1 Patients
Between June 1996 and October 1997, 18 patients underwent minimally invasive aortic valve procedures at our unit. Aortic valve replacement (AVR) was carried out in 17 patients and valve repair in one patient. The patients ranged in age from 42 to 86 years (mean 64 years); there were ten women and eight men. Eight of the AVR patients received mechanical valves and nine received bioprosthetic valves. The annulus was enlarged with an autologous pericardial patch or a Dacron patch in three patients. Subvalvular myocardial resection was performed in one patient, and reduction-plasty of a dilated ascending aorta was performed in one patient. Mean aortic cross-clamp time was 80 min (range 41–125), mean cardiopulmonary bypass time was 104 min (range 55–164), and mean operation time was 209 min (range 115–295).
2.2 General principles
Patients were anesthetized in the supine position and tracheally intubated with a single lumen tube. External defibrillation pads were placed on the chest, which was prepared and draped as for a regular sternotomy. Temporary right ventricular pacemaker wires were applied before the aortic cross-clamp was removed. Standard hemodynamic monitoring was used. Transesophageal echocardiography (TEE) was an essential part of the procedure.
2.3 Incisions
Three types of incision were used: right parasternal minithoracotomy without rib resection (three cases), upper ministernotomy (l1 cases), and transverse sternotomy (four cases).
2.3.1 Right parasternal minithoracotomy
A 10-cm-long skin incision was made along the right sternal border (Fig. 1
A). The pectoral muscle was divided, exposing the third and fourth costal cartilages. The pleura was entered through the fourth intercostal space and the right internal thoracic artery and vein ligated and divided. The third and fourth costal cartilages were divided medially close to the sternum and laterally at the junction to the bony rib (Fig. 1B) to create a vascularized flap of the chest wall (Fig. 4A). The flap was turned into the thorax and a self-retaining retractor was inserted in the wound (Fig. 2
). The pericardium was opened longitudinally and suspended with multiple stay sutures attached under tension to the wound edge. In this way the aortic root and the right atrium became well exposed. The right atrium was cannulated in the usual manner, using a regular two-stage venous cannula (Figs. 2 and 3
). The right femoral artery was electively cannulated in the groin.
View larger version (67K):
[in this window]
[in a new window]
|
Fig. 1. Skin (A) and chest (B) incisions used for parasternal minithoracotomy. The chest was entered through the fourth intercostal space. The right internal thoracic artery and vein were ligated. The third and fourth costal cartilages were separated from sternum and from the bony ribs, creating a vascularized flap of the chest wall (`trapdoor'), which was turned into the chest during the procedure.
|
|
View larger version (99K):
[in this window]
[in a new window]
|
Fig. 4. (A,B) After completion of the valve procedure, the chest wall flap was sutured back in place using double loops of monofilament resorbable sutures for the ribs. (C) Cosmetic appearance of the incision 6 months after surgery.
|
|
View larger version (110K):
[in this window]
[in a new window]
|
Fig. 2. Surgeon's view of the right atrium and the aortic root. The right atrium was cannulated by a regular two-stage cannula.
|
|
View larger version (113K):
[in this window]
[in a new window]
|
Fig. 3. Aortic valve replacement using a St. Jude Medical valve prosthesis. The left ventricle was vented by a thin plastic cannula through the prosthesis. In this case the aortic annulus was enlarged by a gelatin-impregnated Dacron patch.
|
|
After start of cardiopulmonary bypass, the pulmonary artery was vented using passive gravity drainage. The aorta was cross-clamped using a Fogarty clamp (it being ideally angled and having sufficiently long jaws to reach the proximal aortic arch from the wound edge). The aortic root was opened by a transverse incision. Stay sutures were placed above each valve commissure and put under tension. This maneuver pulled the aortic root into the operative field and considerably enhanced exposure [1, 6, 7]. The aortic valve replacement or repair was carried out in the usual fashion (Fig. 3).
After completion of the valve procedure, the chest wall flap was sutured back in place using strong monofilament resorbable sutures (PDS®, Ethicon) for the ribs (Fig. 4
). Then the pectoral muscles were sutured back to the ribs and the skin incision closed with intracuticular sutures. Fig. 4C shows the appearance of the incision 6 months after operation.
2.3.2 Upper ministernotomy
A 10–12-cm-long skin incision was made over the sternum from the jugulum to the level of the fourth intercostal space (Fig. 5
A). A sternotomy was made in the same line, but the bone was cut horizontally or slightly angled at the fourth intercostal space (Fig. 5B). Initially we tried to divide only the right part of the sternum as advocated by Gundry [2], but this was found not to give adequate exposure and also increased the risk of damaging the right internal thoracic artery. The pericardium was incised, exposing the aorta and the right atrium. After heparinization the heart was cannulated in the usual manner (Fig. 6
A). Fig. 6B shows the appearance of this incision 10 days after surgery.
View larger version (67K):
[in this window]
[in a new window]
|
Fig. 5. Skin (A) and sternal (B) incisions for upper ministernotomy. The upper two-thirds of the sternum was split longitudinally in the midline. The sternum was then transected transversely at the level of the fourth intercostal space.
|
|
View larger version (138K):
[in this window]
[in a new window]
|
Fig. 6. (A) Surgeon's view of upper ministernotomy (first case where the skin incision became a bit long). The right atrium was cannulated with a two-stage cannula (right) and the aorta with a regular aortic arch cannula (left). (B) Cosmetic appearance of an upper ministernotomy 10 days after surgery.
|
|
2.3.3 Transverse sternotomy
With the transverse sternotomy, a 10-cm-long transverse skin incision was made over the sternum at the third intercostal space (Fig. 7
A). Both the right and left internal thoracic arteries were ligated and divided. The sternum was divided. A small Finochetti retractor was used to separate the sternum ends (Fig. 8
). The pericardium was opened longitudinally and the edges suspended with stay sutures. With this incision it was possible to cannulate both the aorta and the right atrium (Fig. 8B). Aortic valve replacement or repair was carried out in the usual fashion. Fig. 8C shows the incision 2 days after surgery.
View larger version (67K):
[in this window]
[in a new window]
|
Fig. 7. Skin (A) and sternal (B) incisions for transverse sternotomy. Both internal thoracic arteries were divided and the sternum transected at the level of the third intercostal space.
|
|
View larger version (101K):
[in this window]
[in a new window]
|
Fig. 8. (A) Anesthesiologist's view from the head of the table of a transverse sternotomy. The sternal edges were separated with a Finochetti retractor and the pericardial edges suspended to the wound edge. The right ventricle was well exposed in this case. (B) The right atrium was cannulated with a two-stage cannula and the aorta with an aortic arch cannula. (C) Cosmetic appearance of transverse sternotomy on the second postoperative day after the chest drains were removed. The temporary pacemaker wires are seen to the right.
|
|
2.4 De-airing of the heart (all approaches)
After closure of the aorta, intracardiac air was evacuated by gradual filling of the heart and continuous venting of the ascending aorta by a large-bore cardioplegia cannula at the highest point. Concomitantly the anesthesiologists inflated the lungs and guided the de-airing procedure by TEE. The surgeon compressed the left ventricle anteriorly, with either the back of the fingers (if it was possible to introduce them in the pericardium) or the back of a long pair of scissors or by external chest compression to force the air into the vented aorta.
|
3. Results
|
|---|
There was no mortality and no complications related to the minimal incision or to the procedure. One patient (5.6%) was reoperated for bleeding. Another patient, an 86-year-old man, suffered a transient hemiparesis that was probably embolic in nature. All patients were discharged from hospital within the usual time (7 days). For safety reasons no attempts were made to discharge them earlier even if they seemed to recover more rapidly than patients who had a full sternotomy. There was no instability or paradoxical movements of the chest wall in any of the patients. The cosmetic appearance of the wounds was satisfactory in all patients.
|
4. Discussion
|
|---|
Our experience with minimal access aortic valve surgery in Linköping is small, but some conclusions may still be drawn. Of the incisions used, the upper ministernotomy seemed to be the best. It was easy to perform and gave adequate exposure, and the internal thoracic arteries were usually not damaged. This incision could also be converted to a regular median sternotomy if required. Another advantage was that the ascending aorta and the beginning of the aortic arch were very accessible and could be scanned by ultrasound in elderly patients. Concomitant procedures on the ascending aorta could also be performed through this incision.
From a cosmetic standpoint, the right parasternal minithoracotomy and the transverse sternotomy were the best. These incisions were not visible in the patients' necks with an open shirt (Fig. 4C and Fig. 8C). The 7 patients operated on with these techniques also seemed to have little pain.
There were no problems with instability of the chest wall. With the parasternal minithoracotomy, the ribs were sutured back in place with monofilament resorbable sutures, and with the transverse sternotomy, in addition to three steel wires in the sternum, we approximated the adjacent third and fourth ribs with double loops of monofilament resorbable sutures to reinforce the sternal osteosynthesis.
A drawback of the parasternal minithoracotomy was that it was difficult to cannulate the ascending aorta. For safety reasons, we preferred to electively cannulate the femoral artery and had no problems. However, we still feel that it is an advantage if all cannulation can be done through the chest incision. One surgical group has been able to routinely cannulate the aorta through the parasternal incision [3]. Cannulation of the right atrium caused problems in only 2 cases where we were not able to steer the tip of the two-stage cannula into the inferior vena cava. We simply cut off the tip of the cannula and introduced its main part into the right atrium, which worked well. Due to difficulties steering the cannula, it would be an advantage to have an angled two-stage cannula. Recently a 90-degree [4]and a 45-degree [5]cannula have been described that would be suitable for this purpose.
A disadvantage of the transverse sternotomy is that both internal thoracic arteries must be sacrificed. We have therefore used this incision only in elderly patients with angiographically normal coronary arteries where a subsequent coronary bypass operation is highly unlikely. Even with the upper ministernotomy, there is a risk of damaging the internal thoracic arteries, particularly the right, if the sternum edges are separated too widely.
One major drawback of minimal incisions is that not all patients can be operated on through them. This is particularly true for patients with concomitant coronary artery disease, where only those with a proximal right coronary artery lesion can have a minimal incision. In our experience the upper ministernotomy ending at the fourth intercostal space gave the best exposure of the proximal right coronary artery. Patients who are massively overweight are not good candidates for any of the minimally invasive procedures. The fat in the mediastinum and the aortic root impairs exposure and makes the operation less safe.
Our usual method of venting the heart during AVR is to use either a left apical vent or a vent through the right superior pulmonary vein. As this was difficult in the present cases, we had to vent through the ascending aorta or the pulmonary artery. The latter method was the easiest but introduced a potential risk of sucking air into the pulmonary artery system, which could cause problems later. We therefore used passive gravity drainage (such as on the venous return) whenever possible, which we felt was safer.
De-airing of the heart after completed valve surgery was more difficult than with median sternotomy. The best technique for de-airing was to fill the left ventricle gradually, tilt the operating table to the left and the feet down, and then expel the air into the aorta by external manual compression of the left chest or by compressing the left ventricle with the end of long scissors. Simultaneously the anesthesiologists inflated the lungs and guided the de-airing procedure by TEE. The most important part of the de-airing procedure was the continuous venting of the ascending aorta with a large-bore needle at the highest point. After release of the aortic cross-clamp, the patient was placed in the Trendelenburg position so that any intracardiac air ejected during the first heart beats was evacuated through the vent or passed to the lower part of the body.
Finally, it is important to recognize that there are several components to an AVR operation, including anesthesia, chest incision, cardiopulmonary bypass, clamping of the aorta, cardioplegia, and valve excision and replacement. It is only the chest incision component that can be improved by minimal access; the other components are the same. Therefore, only moderate benefits can be expected in the early postoperative period. These benefits may only be apparent during the later phase of recovery. Defining the true value of minimal access surgery requires carefully controlled studies.
In conclusion, minimal access aortic valve surgery can be carried out by many different approaches. Of the three incisions used, the upper ministernotomy seemed to be the safest and most useful. Through this incision, the ascending aorta and the right atrium could be cannulated and concomitant procedures involving the ascending aorta performed. No special surgical instrumentation was needed; however, de-airing was more difficult through minimal access incisions than with regular sternotomy.
|
Conference discussion
|
|---|
Dr R. Wallace
(Washington, DC, USA): I saw that four of your aortic valve replacements required outflow tract patches. That seems to be a high incidence. Could you comment on whether this approach affects that at all?
Dr C. Olin
: Many of these patients are women, and women often have small aortic roots. Our policy is to implant as large a valve as possible without any significant residual pressure gradient, because the long-term prognosis benefits from low gradients. Otherwise regression of left ventricular hypertrophy will not occur. If we cannot implant a valve sufficiently large for the patient's body size and activity demands, we enlarge the aortic annulus.
Dr R. Wallace
: Was it difficult to reach the posterior annulus, and what kind of patch did you use?
Dr C. Olin
: No. I think it was easy to reach. And usually we use the patient's own pericardium, which we soak in 1% glutaraldehyde solution for 5 min, before the enlargement procedure. This enlargement procedure is very easy and takes about 8 min. It thus might prolong the aortic cross-clamping time a little but is worth the effort. We use it in about 20% of our cases with small aortic roots.
Dr J. Craver
(Atlanta, GA, USA): Aortic valve minimally invasive surgery is here to stay. I would agree that rehabilitation is significantly shorter. However, I do not think early pain is any different, and I think the hospital stay is about the same. Of your three demonstrated types, which do you prefer? With the trapdoor right parasternal, do you cannulate femorally? I did not see an aortic cannula on that picture. How do you vent them? And have you ventured to do much larger operations, such as a Bentall, through similar procedures? In older patients, who comprise a significantly increasing component of aortic valve patients and in whom the aorta is occasionally quite elongated and dependent, do you make an effort preoperatively to assess the location before doing the transverse component of any aspect of the approach?
Dr C. Olin
: The partial median sternotomy causes more pain than the other two incisions. The trapdoor incision causes very little pain, and we have had good experience with it, including some mitral valve surgery. The transverse incision also causes very little pain; some people have had problems with sternal instability, but we have used double-loop resorbable monofilament sutures in addition to the sternal wires and have achieved very stable fixation without any problems. My personal preference depends on the situation. If it is a difficult case involving complicated surgery on the ascending aorta or the aortic root, I would do a partial median sternotomy. If it is a lovely, young woman with isolate AVR, I would do a trapdoor incision. The drawback of this incision is that it is difficult to cannulate the aorta through it. We have therefore electively cannulated the femoral artery in these cases. We have had several patients with Bechterev's disease where they have not been able to bend their necks backwards, and in these cases the transverse incision is excellent because you do not have to go up in the neck. So you have to select the cases. We have not yet performed extensive aortic root reconstructions for aortic root ectasia using this technique, even if it might be feasible; in those cases we have done a conventional sternotomy. We have tried to locate the level of the aortic root by TEE but have not yet evaluated the technique.
Mr P. Deverall
(London, UK): One of our responsibilities is to train the surgeons of the future. When a resident is being taught, which technique do you select to teach him to do an aortic valve replacement safely and why?
Dr C. Olin
: These operations are not popular with residents in training. Those cases have been done so far by the consultants because there is a learning curve and the consultants want to get comfortable with the technique themselves first. We have not yet tried to train our residents, but if I were to teach the operation to a resident, I would start with the upper partial sternotomy.
Dr A. Juffé
(La Coruña, Spain): We have done 25 cases using the Cosgrove transverse incision with no femoral cannulation and excellent results. But we had sternal dehiscence in 3 out of the first 10 cases, so we now use a double-loop suture. What is your incidence of sternal dehiscence?
Dr C. Olin
: We have only done 4 cases of transverse sternotomy but have not had any problems with sternal instability. I think the extra double-loop resorbable sutures around the adjacent ribs are excellent for stability and can be used for a lot of other sternal problems.
Dr N. Al-Khaja
(Dubai, United Arab Emirates): It seems that it has become easy to do this procedure with aortic valve replacement. Have you shifted totally to this minimally invasive aortic valve replacement or are you still doing the conventional type and why?
Dr C. Olin
: We have not yet shifted to it because the isolated AVRs are usually performed by senior residents. And with the healthcare system in Sweden, there is no real pressure from the patient. There is also no pressure yet from the cardiologist because the real benefits of the method and the risks involved have not yet scientifically been established.
|
Footnotes
|
|---|
1 Presented at the Sulzer Carbomedics Sixth International Clinical Symposium, Copenhagen, Denmark, 27 September 1997. 
|
References
|
|---|
- Cosgrove DM, Sabik JF. Minimally invasive approach for aortic valve operations. Ann Thorac Surg 1996;62:596-597.[Abstract/Free Full Text]
- Gundry RS. Ministernotomy approach to aortic valve repair/replacement. The American Association for Thoracic Surgery, 77th Annual Meeting, Presentation at the Adult Cardiac Surgery Symposium, May 4, 1997..
- Minale C, Reifschneider HJ, Schmitz E, Uckmann FP. Single access for minimally invasive aortic valve replacement. Ann Thorac Surg 1997;64:120-123.[Abstract/Free Full Text]
- Bugge M, Lepore V, Dahlin A. The `90 degree bent' two-stage venous cannula. Eur J Cardio-thorac Surg 1995;9:526-527.[Abstract]
- Lawrence DR, Desai JB. Forty-five-degree two-stage venous cannula: advantages over standard two-stage venous cannulation. Ann Thorac Surg 1997;63:253-254.[Abstract/Free Full Text]
- Konertz W, Waldenberger F, Schmulzler M, Ritter J, Jiauoli L. Minimal access valve surgery through superior partial sternotomy: a preliminary study. J Heart Valve Dis 1996;5:638-640.[Medline]
- Szerafin EJ, Jaber O, Horváth G, Olin C, Péterffy Á. Mini-sternotomy for aortic valve surgery. Acta Chir Hung 1997;36:352-355.[Medline]
Top
Abstract
1. Introduction
