|
|
||||||||
a Clinic for Cardiovascular Surgery, Zurich, Switzerland
b Institute of Diagnostic Radiology, University Hospital Zurich, Zurich, Switzerland
c Cardiovascular Center, University Hospital Zurich, Zurich, Switzerland
Accepted for publication August 6, 2009.
* Address correspondence to Dr Plass, Clinic for Cardiovascular Surgery Ramistr. 100, Zurich, 8091, Switzerland (Email: andre.plass{at}usz.ch).
| Abstract |
|---|
|
|
|---|
Methods: From March 2006 to June 2008, 172 patients (113 men; mean age, 71 ± 12 years) were scheduled for MIAVR (6- to 7-cm incision). Multislice computed tomography (MSCT) imaging was used for surgical planning in 139. Aortic cannulation/clamping were performed through a right-sided minithoracotomy and venous cannulation percutaneously through the groin. For obtaining optimal intercostal space (ICS) distances between the incision to the aorta and cardiac structures, 2- and 3-dimensional MSCT images were evaluated.
Results: Operations were done in 171 patients. MIAVR was successfully performed in 160 (94%). Six patients underwent a conventional operation due to adhesions in 4, small diameter of aortic annulus (17 mm) in 1, and concomitant coronary artery disease in 1. One patient was considered nonoperable. After CT-planning choice of second ICS in 17%, third in 81%, and fourth in 1%. Five conversions to sternotomy were necessary. Intraoperative and postoperative complications occurred in 20 patients, including 1 death. Overall cardiopulmonary bypass was 158 ± 41 min and cross-clamp time was 107 ± 26 min. No blood products in 43% of MIAVR patients. Mean hospital length of stay was 10 ± 3 days.
Conclusions: MIAVR demonstrates excellent results. A considerably reduced complication rate in the course was noted. MSCT for preoperative planning is helpful for an improved mental preparation and for an accurate surgical strategy, including optimal access.
| Introduction |
|---|
|
|
|---|
Since 2006 we have performed minimally invasive aortic valve replacement (MIAVR) through a lateral right minithoracotomy. This surgical approach limits the view of the surgeons and the space of the operating field, which results in a more challenging procedure. An optimal mental preparation can support the surgeon to better overcome the difficulties caused by the new modalities of minimally invasive surgery.
Multislice computed tomography (MSCT) has already been used successfully as part of the clinical routine in cardiac operations for many different tasks, including detailed illustration of the anatomy of the heart, examination of coronary arteries, coronary bypass grafts, and heart valves [4–9]. This imaging tool can be used for diagnostic and postoperative quality control as well as for preoperative planning. Preoperative planning with MSCT has shown to be a helpful tool in reoperations and minimally invasive cardiac operations [10, 11], demonstrating an improved orientation in the operating field and potentially preventing surgical errors.
In this article we report our approach to MIAVR and the associated complication rate as well as the short-term outcome. In addition, we illustrate the experiences and the advantages of preoperative planning with MSCT specifically designed for this surgical technique.
| Patients and Methods |
|---|
|
|
|---|
Patients
From March 2006 to June 2008, elective MIAVR was planned in 172 consecutive patients (109 men and 63 women; mean age, 70 ± 12 years) with isolated aortic valve disease, comprising 79% with aortic stenosis, 7% with regurgitation, and 14% with combined aortic valve disease. The patients had a European System for Cardiac Operative Risk Evaluation (EuroSCORE) of 6% ± 4% (range, 2% to 18%). Demographic data for these patients are presented in Table 1.
|
|
|
After adequate electromechanical cardiac arrest, the aortic valve is exposed through an oblique/transverse aortotomy and 3 commissural retracting sutures. The native valve is excised and the annulus thoroughly decalcified. Prothesis size is chosen after sizing the annulus with the appropriate valve sizers. Valve implantation is performed in supraannular position using a noneverting technique with 2-0 Ethibond pledgeted sutures (Ethicon, Norderstedt, Germany) with minimally invasive surgical instruments (ValveGate, Geister, Germany). The patient is rewarmed and the aortotomy is closed.
Before the aorta is unclamped, a ventricular pacing wire is placed on the right ventricle and brought out through the skin. Then the cross clamp is removed. If no adequate cardiac rhythm is present after declamping, external defibrillation is applied through externally placed defibrillation pads. The patient is weaned from CPB and all cannulas are removed. Protamine is given and a chest tube is inserted into the right pleural space. The ribs are approximated and the pectoral muscle is readapted. Subcutaneous and intracutaneous sutures are applied for wound closure.
MSCT Imaging
For preoperative planning patients underwent imaging with a dual-source CT system (Siemens Healthcare, Erlangen, Germany). The patients were examined with their arms tucked to the chest wall according to the position of the patient in the operating room. An ancillary electrocardiograph (ECG) electrode was affixed on the right side of the chest approximately one hand-width inferior to the clavicle to provide a landmark for orientation on the skin and identification of the incision point. No intravenous β-blockers were given before the scan.
Patients first underwent unenhanced CT for the quantification of calcifications before the contrast-enhanced scan. A bolus of 70 to 100 mL (adjusted to the individual body weight and scan time) of nonionic, iodinated contrast material (Ultravist 370, 370 mg iodine/mL, Bayer Schering Pharma, Leverkusen, Germany), followed by a 40-mL saline flush, was injected into an antecubital vein for contrast-enhanced CT. All scans were performed in ECG-triggered mode.
The reconstructions of unenhanced CT scans were performed with a mono-segment mode using 3-mm-thick nonoverlapping sections (reconstruction kernel B35). The Agatston score was calculated from the unenhanced CT data with a detection threshold of 130 HU by using semiautomated software (Calcium Scoring CT, Siemens). Contrast-enhanced CT scans were reconstructed in a mono-segment mode with an overlapping slice thickness of 0.75 mm (increment, 0.5 mm). All images were transferred to an external workstation (Multi-Modality Workplace, Siemens) for further analyses.
CT Data Analysis and Preoperative Planning Protocol
After data acquisition, the complete image stack was transferred to an external workstation (Leonardo, Siemens) equipped with dedicated cardiac postprocessing software (InSpace, Siemens). Image interpretation was performed from axial-source images, multiplanar reformations, and 3-dimensional (3D) volume renderings.
First, the location and morphology of the surgical regions of interest was determined by performing 3D measurements (Table 2). For distance measurements, an electronic caliper was laid between both points after identifying the start and end point on the axial-source images. The diameter of the aortic valve annulus was evaluated on double-oblique multiplanar reformations parallel and perpendicular to the annulus. The length of the ascending aorta was assessed with a curved 3D caliper applied on the surface of the aortic volume renderings, and the calcifications of the aortic valve and the ascending aorta were evaluated (Figs 3, 4).
|
|
|
Surgical Data Analysis
MIAVR was assessed with regard to various intraoperative and perioperative data: CPB time, aortic cross-clamp time, conversion rate, complications (reexploration, injuries, neurologic events, wound infection, transfusion rate of blood products), length of intensive care unit (ICU) stay, length of hospitalization, and 30-day mortality.
Statistical Analysis
Continuous variables were expressed as mean ± standard deviation and categoric variables were expressed as frequencies or percentages. For comparison of the continuous variables, the t test was used. A value of p < 0.05 was regarded as statistically significant.
| Results |
|---|
|
|
|---|
After further preoperative examinations, 1 patient with multiple comorbidities was considered inoperable. Six other patients underwent conventional AVR through a median sternotomy because of strong adhesions in 4 patients, additional coronary artery bypass grafting (CABG) in 1, and unusual anatomy with small annulus (17 mm) in 1 patient.
Finally, 165 patients were scheduled for MIAVR. After the initial lateral thoracotomy, 5 patients (3%) underwent intraoperative conversion to a median sternotomy: in 2 patients due to unexpected strong adhesions in the pleural space, in 1 patient for additional CABG because of an accidental occlusion of the right coronary artery after valve implantation, in 1 patient because of acute aortic dissection, and in 1 patient because of injury to the right ventricle during venous cannulation. In 4 of the 5 converted patients, the procedure was uneventful after conversion except for the patient with the acute aortic dissection, who died.
For optimal surgical access according to the measurements in the MSCT, the second ICS was chosen in 17%, the third ICS was used in 81%, and the fourth ICS was used in 1%. In the 26 patients who did not undergo CT imaging, the third ICS was always chosen. The distance was 6 ± 1 cm from the skin to the nearest point of the aorta for cannulation and 8 ± 1 cm to the aortic valve for prosthesis implantation.
In 60% of the patients, calcified plaques were shown on the ascending aorta. This information from the CT scan influenced the aortic cannulation site in 3 patients: 1 patient received a short-tip cannula instead of the usually used long-tip cannula, and 2 patients underwent cannulation of the subclavian artery. The volumetric calcium score of the aorta for all assessed patients was a mean of 415 ± 1016 mm3. The mean values of the different measurements of the standardized protocol for lateral thoracotomy for AVR are listed in Table 2.
In this study, 126 biologic valves (Mosaic Standard and Mosaic Ultra, Medtronic, Minneapolis, MN) and 34 mechanical valves (Regent, St. Jude Medical, Minneapolis, MN) were successfully implanted in a minimally invasive way. The valve sizes used were in accordance with the measurements in CT and intraoperative echo imaging.
Postoperative complications occurred in 9% of patients (15 of 160) operated on successfully with MIAVR, including 4% reexplorations for bleeding, 4% neurologic events, 1% rhythm disorders requiring a permanent pacemaker, and 0.6% pericardial drainage during hospitalization. No wound infections (0%) occurred. The 30-day mortality was 2% (3 of 160).
Mean intraoperative and postoperative blood product transfusion requirements were 2 ± 3 U erythrocyte concentrates and 0.3 ± 0.7 U thrombocyte concentrates (1 U = 300 mL). No blood products were required in 68 of the MIVAR patients (43%).
The ICU length of stay was 2 ± 2 days, whereas 106 patients (66%) only stayed 1 day. The mean hospital length of stay was 10 ± 3 days (Table 3).
|
| Comment |
|---|
|
|
|---|
Postoperative recovery in the ICU, at the ward, and later on during the rehabilitation phase was clearly much improved in MIAVR patients compared with patients after sternotomy, mainly due to less pain. Regarding the operating technique, the aortic cannulation is one of the most important features that can be performed directly through the thoracotomy instead of by peripheral arterial cannulation in the groin. We believe that peripheral arterial cannulation should be avoided in AVR operations to prevent neurologic complications because the aorta in these patients is rarely normal. In mitral valve repair, however, peripheral arterial cannulation is much safer because of the lack of aortic calcifications in most patients. Only 4 patients in our study had proven neurologic events, which is less then 2% of all patients.
Since the introduction of the MIAVR program in our institution, a learning curve has been noted regarding complications and length of CBP and aortic cross-clamp time. Although these times are still longer compared with conventional procedures, the most time-consuming part of the operation is placing the sutures and tying the knots, especially if a knot pusher is necessary. A significant step forward to reducing the aortic cross-clamp time and CPB time would be if sutureless valves could be used. We believe that using such valve prostheses would reduce the cross-clamp time to less than 20 minutes, making this procedure a real alternative to transcatheter valves in the future.
Imaging and Preoperative Planning
Imaging techniques are increasingly more important, which can be defined in four different pillars: pillar 1: diagnostic tool; pillar 2: preoperative planning and simulation; pillar 3: intraoperative navigation and controlling; and pillar 4: postoperative follow-up and quality control.
Standard diagnostic technique in cardiac operations, such as coronary angiography for CABG and echocardiography for valve repair or replacement (pillar 1) have also always been used for preoperative planning (pillar 2).
Optimal preoperative planning is comparable with a preparation of a rally car driver before a competition. Equally, a surgeon will not be able to finish an operation in the same manner as he or she would being fully aware of the specific anatomy and the surgical consequences (ie, access through the ICS, cannulation site, and type of cannula). The more complex a surgical procedure, the more important is the additional information that can be gathered with MSCT imaging [5, 11]. However, in which situation and under what circumstances they should be used depends on the individual surgeon and is clearly debatable.
Preoperative planning with MSCT was a helping factor practically and mentally to prepare the surgeon for the MIAVR procedures. However, to be able to really prove if preoperative planning will lead to a reduction of the operating time and complication rate, a prospective randomized comparative trial would be necessary, which is difficult to perform.
The preoperatively placed skin markers on the second and third rip were very helpful, especially in obese patients. With these markers the surgeon was able to easily identify the second or third ICS, which he deemed most suitable for each patient individually. The chosen ICS, according to the preoperatively acquired MSCT, was optimal in most of the patients. The measurements from the second or the third ICS were either better for cannulation or better for replacing the valve. In only a few cases was the ICS changed intraoperatively before cannulation, which was more common in the patients with no preoperative CT scan.
ECG gating was used to synchronize all MSCT examinations to the heart motion. The anatomy of the heart was therefore depicted under beating conditions. All open heart procedures, however, are performed on the arrested heart, which might lead to slight differences in the measurements compared with a beating heart. However, not only are the absolute values important, but also the relationship of the different measurements to the region of interest is also essential.
Regarding the thoracotomy or sternotomy, we experienced three incidences in which, in retrospect, the performed thoracotomies were debatable. In 1 patient, the distances to the aortic valve were doable, yet at the limits; however, access from the incision to the region of interest exhibited an extremely sharp angle that limited the maneuverability of the long-shafted instruments. Furthermore, adhesions are still very problematic. Not only the detection but also the grading of pleural adhesions is very difficult. In 2 patients the surgical procedure was so difficult that the lateral thoracotomy had to be converted into a median sternotomy. Therefore, the goal will be to accurately evaluate adhesions in the pleura and pericardium to make a decision whether these adhesions will influence the operation.
Preoperative planning is not only the evaluation of different parameters, including the illustration of the anatomy for minimally invasive surgery and optimizing the surgical strategy itself, but should also support the decision of whether a minimally invasive operation is possible. Criteria that are very difficult to describe are an atypical anatomy and individual distances, especially in obese patients. In these cases, only the surgeon can decide if he or she feels comfortable performing a minimally invasive procedure. To train for these individual circumstances, the development of a virtual reality interactive preoperative simulation of the procedure individual to the patient would be very helpful.
In conclusion, MIAVR demonstrates excellent short-term results. Preoperative planning with MSCT leads to an improved mental preparation and to an efficient and accurate surgical strategy including the choice of the optimal intercostal space.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Murzi, A. G. Cerillo, S. Bevilacqua, D. Gilmanov, P. Farneti, and M. Glauber Traversing the learning curve in minimally invasive heart valve surgery: a cumulative analysis of an individual surgeon's experience with a right minithoracotomy approach for aortic valve replacement Eur J Cardiothorac Surg, January 9, 2012; (2012) ezr230v1. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Bruschi, F. De Marco, P. Fratto, J. Oreglia, P. Colombo, L. Botta, S. Klugmann, and L. Martinelli Alternative approaches for trans-catheter self-expanding aortic bioprosthetic valves implantation: single-center experience Eur J Cardiothorac Surg, June 1, 2011; 39(6): e151 - e158. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Breitenbach, G. Wimmer-Greinecker, L. A. Bockeria, J. Sadowski, C. Schmitz, B. Kapelak, K. Bartus, R. Muratov, and W. Harringer Sutureless aortic valve replacement with the Trilogy Aortic Valve System: Multicenter experience J. Thorac. Cardiovasc. Surg., October 1, 2010; 140(4): 878 - 884.e1. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Iribarne, R. Easterwood, M. J. Russo, J. Yang, F. H. Cheema, C. R. Smith, and M. Argenziano Long-Term Outcomes With a Minimally Invasive Approach for Resection of Cardiac Masses Ann. Thorac. Surg., October 1, 2010; 90(4): 1251 - 1255. [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 |