|
|
||||||||
Department of Cardiothoracic Surgery, Lenox Hill Hospital, New York, New York
Accepted for publication January 23, 2008.
* Address correspondence to Dr Loulmet, Department of Cardiothoracic Surgery, Lenox Hill Hospital, 130 E 77th Street, New York, NY, 10075 (Email: loulmetd{at}aol.com).
Presented at the Poster Session of the Forty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 29–31, 2007.
| Abstract |
|---|
|
|
|---|
Methods: From March 2000 to September 2006, 100 patients underwent MV repair (n = 42), MV annuloplasty (n = 28), MV replacement (n = 18), atrial septal defect closure (n = 10), tricuspid valve repair (n = 1), and left atrial myxoma excision (n = 1). A modified maze procedure (n = 4) or left minimally invasive direct coronary bypass grafting (MIDCABG) (n = 2) was combined in six cases. The mean age was 57 ± 11 years (range, 22 to 89); 27 patients were in New York Heart Association (NYHA) class III or IV; 24 cases were first or second time reoperations; 20 patients had a left ventricular ejection fraction of less than 0.3. All the operations were carried out on the fibrillating heart without cross-clamping the aorta through a right minithoracotomy using peripheral cannulation.
Results: Mean fibrillation time was 73 ± 31 minutes (range, 10 to 198 minutes). There was no conversion to sternotomy. Postoperative inotropic support was needed in 20 cases. One patient who underwent a third time reoperation died within 30 days of mesenteric ischemia (hospital mortality = 1%). Complications were the following: four reoperations for bleeding (4%); two strokes (2%). Postoperative median hospital length of stay was five days (range, 2 to 58 days). None of the patients has required MV reoperation after hospital discharge. Follow-up was complete. All survivors were in NYHA class I or II.
Conclusions: Ventricular fibrillation simplifies less invasive intracardiac procedures and carries lower complication rates and perioperative mortality compared with conventional surgery.
| Introduction |
|---|
|
|
|---|
| Patients and Methods |
|---|
|
|
|---|
Patient Selection
During the study period 1,278 patients underwent mitral valve surgery at our institution. At the beginning, the technique described in this paper was only indicated in patients with prior coronary bypass surgery and patent grafts in need of an isolated mitral valve annuloplasty. For these patients, the changes we made compared with the techniques reported in the literature were the use of anterior minithoracotomy (opposed to a regular thoracotomy) and peripheral cannulation (opposed to central cannulation). These changes were made based on our prior experience with less-invasive mitral valve surgery. We progressively extended the indications of this technique to non-redo patients in need of an isolated mitral valve annuloplasty or replacement or atrial septal defect (ASD) repair. Later on, we involved patients presenting with an isolated prolapsus of the posterior leaflet. Patients with a prolapsus of the anterior leaflet demonstrated on preoperative or intraoperative transesophageal echocardiography (TEE) were not candidates for this technique. Selection has also been, all along, dependent on surgeon's preference.
Surgical Techniques
Blood pressure was monitored through the left radial artery. A double-lumen endotracheal tube was used for left lung ventilation and its position checked with bronchoscopy after final positioning of the patient. An endopulmonary vent was placed through the right jugular vein. It was removed and replaced with a Swan-Ganz catheter at the end of the case in patients with significant left ventricular dysfunction. Transesophageal echocardiography was used in all cases. External defibrillation patches were placed before draping. The patient was placed in lateral decubitus with a slight elevation of the right side.
The right chest was entered in all cases through a less than 2.5-inch anterior thoracotomy performed in the fourth intercostal space. The skin incision was made below the right breast in the mammary fold for women and above the right nipple for men. A soft tissue retractor (Cardiovation; Ethicon Inc, Somerville, NJ) and an intercostal retractor (Estech; Estech Corporation, Danville, CA) were used for limited opening. A transparietal suture was placed to lower the right hemidiaphragm. Carbon dioxide was continuously insufflated into the right pleural cavity using a transparietal needle (2 liters per minute flow).
After heparinization, the left femoral artery was exposed (the right femoral artery had often been used for catheterization the day before) and cannulated with a direct Seldinger technique for arterial return. Axillary cannulation was used in three patients with peripheral atherosclerotic disease. Venous drainage was provided by a central cannula (Cardiovation) placed in the right atrium through the right femoral vein using a percutaneous Seldinger technique. The tip of the cannula was placed in the superior vena cava. Percutaneous right jugular cannulation was added if the right atrium needed to be opened and the vena cave snared.
Cardiopulmonary bypass (CPB) was instituted and the body temperature allowed drifting not lower than 34°C. We did not use active cooling. Mean arterial pressure during perfusion was kept above 60 mm Hg. The surgery was performed under direct vision using endoscopic instruments (Estech or Cardiovation) and a knot pusher (Cardiovation). The pericardium was opened longitudinally above the right phrenic nerve. Two pacing wires were used to induce and maintain ventricular fibrillation with one placed on the diaphragm and the other on the right ventricle. At the end of the procedure, defibrillation was undertaken with external defibrillation pads when TEE demonstrated adequate removal of air. Both lungs were kept inflated in order to maximize conductivity at the time of cardioversion.
Mitral Valve Patients
After the induction of ventricular fibrillation, the left atrium was entered through the Sondergard groove. The mitral valve was exposed with the left atrial retractor (Cardiovation) held by a mechanical arm (Martin Arm) fixed to the left side of the table. In order to minimize aortic valve regurgitation, the left atrial retractor was placed in an oblique position, pushing the aortic root toward the left rather than retracting toward the sternum. Two flexible vents were used with one placed in the left inferior pulmonary vein and the other in the left ventricle. The placement of annulus stitches at the level of the anterior commissure can be difficult due to the bulging of the pressurized aortic root into the left atrium. Turning the pump off for a period of less than 15 seconds at the time of stitch placement facilitated the exposure. Removal of air from the left cavities was performed without venting the aorta or placing the patient in the Trendelenburg position. As a first step, air was removed from the left atrium through the atriotomy suture line with filling the heart and ventilating both lungs. As a second step, air was removed from the left ventricle with a transmitral vent placed at the apex and turned on only after air had been removed from the left atrium and the atriotomy suture line tightened.
Tricuspid Valve Patients
Both vena cavae were snared and the right atrium was entered through a longitudinal incision on the beating heart. The tricuspid valve was exposed with stay sutures retracting the edges of the atriotomy. Two flexible vents were used with one placed in the coronary sinus and the other in the right ventricle. The beating heart technique allowed monitoring A-V conduction when the annulus stitches were placed in the region of the A-V node.
ASD Patients
Both vena cavae were snared and the right atrium was entered through a vertical incision after ventricular fibrillation was induced. Two flexible vents were used with one placed in the coronary sinus and the other in the left atrium. Air was removed from the left atrium through the patch suture line before tying the knot: the left cavities were filled by increasing pump flow, which increased pulmonary vein return into the left atrium while both lungs were inflated. Then the heart was defibrillated and the right atrium closed on the beating heart.
| Results |
|---|
|
|
|---|
|
|
|
| Comment |
|---|
|
|
|---|
Many publications have advocated the use of a standard right thoracotomy approach in combination with induced ventricular fibrillation (or a beating heart technique) for mitral valve surgery [1–8]. In all the reports this technique was used only for reoperative cases whether the patients had had prior CABG [1, 3, 5], valve surgery [2, 6, 8], or both [4, 7]. These studies have demonstrated good results in terms of perioperative morbidity and mortality. In four series [3, 4, 6, 8], comparison was made with a control group in which patients were operated with a standard technique: the data demonstrated that the use of a right thoracotomy approach, in combination with induced fibrillation, decreased coronary artery graft injury [3], transfusion requirements [3, 6, 8], postoperative inotropic requirements [6], incidence of early reoperations [8], and hospital mortality [4].
We modified this technique by limiting the size of the incision to 2.5 inches and by using peripheral arterial and venous cannulation in all instances. We have used this technique for both reoperative and non-reoperative patients. The avoidance of aortic clamping and cardioplegia delivery, whenever it is possible, simplifies the procedure. However, the efficacy of this technique in terms of quality of exposure, myocardial protection, and safety of removal of air can be questioned.
With this technique, exposure of the mitral valve can be limited by three factors: (1) the bulging of the pressurized aortic root into the left atrium hiding the A1 region; (2) some degree of rotation of the mitral valve orifice toward the left in reoperative cases due to pericardial adhesions; and (3) some aortic valve regurgitation caused by exaggerated retraction. Turning the pump off for a very short time decreases the prominence of the aortic root into the left atrium and facilitates stitch placement in the A1 segment of the mitral annulus without allowing air entry in the aortic root. In reoperations, making the minithoracotomy more lateral compensates for the displacement of the mitral orifice. Aortic valve regurgitation can be avoided by placing the left atrial retractor in an oblique position and by pushing the interatrial septum with the aortic root toward the left rather than retracting vertically toward the sternum, which would distort the aortic annulus. Adequate exposure of the mitral valve can be obtained in most cases by using these techniques. Increasing the size of the incision does not provide better exposure. However, we do not recommend this approach in patients needing the correction of anterior leaflet prolapse because complex work on the papillary muscles is made easier when the left ventricle is arrested and relaxed.
Subendocardial perfusion is dependent on perfusion pressure when the ventricles are fibrillating. For that reason the perfusion pressure was maintained above 60 mm Hg. The systemic temperature was allowed to drift to 34°C without using active cooling. Recovery of myocardial function after defibrillation was good as far as we could judge based on rapid normalization of the electrocardiographic tracings and low inotropic requirements, especially in patients with left ventricular dysfunction. This contrasted with our experience with other minimally invasive techniques in which the aorta was clamped with an endoballoon or a transthoracic clamp. We believe that maintaining constant low myocardial temperature in the closed chest environment, with cardioplegia delivery, is difficult unless the body temperature is significantly decreased. On the other hand, endoballoon or transthoracic aortic clamping carries some limitations in reoperative patients with patent coronary artery bypass grafts. Myocardial fibrillation as a technique for arresting the heart during surgery has been compared with standard cardioplegia technique with respect to the postoperative level of creatinine kinase and its myocardial fraction: those were found to be significantly higher in patients operated with cardioplegia [11].
Removal of air was performed without the use of any vent placed on the ascending aorta. We only used retrograde removal of air. Carbon dioxide insufflation into the chest cavity seemed to be efficacious at replacing air inside the heart cavities after cardiotomy. This resulted consistently in a more rapid dissolution of bubbles detected with intraoperative TEE compared with the standard sternotomy procedures. The stroke risk was decreased because of absence of manipulation of the ascending aorta, the use of axillary cannulation in patients with peripheral vascular disease, and a reliable technique for removal of air. We acknowledge the limitations of this retrospective study consisting of a short series of nonconsecutive selected patients presenting with a vast range of pathologies, with no control group for comparison.
In conclusion, we do agree with publications reporting good results obtained with reoperative intracardiac procedures performed on high-risk patients through a right thoracotomy on the fibrillating heart. We adapted this technique on the basis of our experience with less invasive cardiac surgery. We used it in both high-risk reoperative patients and regular non-redo patients with good results. We recommend this approach not only for high-risk reoperative mitral valve patients but also for standard mitral valve repair cases in which the anterior leaflet is not involved, for tricuspid valve surgery, and for other simple intracardiac procedures such as tumor removal or septal defect closure.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
H. H. Dave, M. Comber, T. Solinger, D. Bettex, A. Dodge-Khatami, and R. Pretre Mid-term results of right axillary incision for the repair of a wide range of congenital cardiac defects Eur. J. Cardiothorac. Surg., May 1, 2009; 35(5): 864 - 870. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. S. McClure, L. H. Cohn, E. Wiegerinck, G. S. Couper, S. F. Aranki, R. M. Bolman III, M. J. Davidson, and F. Y. Chen Early and late outcomes in minimally invasive mitral valve repair: an eleven-year experience in 707 patients. J. Thorac. Cardiovasc. Surg., January 1, 2009; 137(1): 70 - 75. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |