The goal of this protocol is to describe in detail the technique of minimally invasive aortic valve replacement through a right anterior mini-thoracotomy and central aortic cannulation. This technique can potentially enhance patients' comfort and, by reducing post-operative morbidity, promote lowering the length of stay and global costs.
Aortic valve stenosis has become the most prevalent valvular heart disease in developed countries, and is due to the aging of these populations. The incidence of the pathology increases with growing age after 65 years. Conventional surgical aortic valve replacement through median sternotomy has been the gold standard of patient care for symptomatic aortic valve stenosis. However, as the risk profile of patients worsens, other therapeutic strategies have been introduced in an attempt to maintain the excellent results obtained by the established surgical treatment. One of these approaches is represented by transcatheter aortic valve implantation. Although the outcomes of high-risk patients undergoing treatment for symptomatic aortic valve stenosis have improved with transcatheter aortic valve replacement, many patients with this condition remain candidates for surgical aortic valve replacement. In order to reduce the surgical trauma in patients who are candidates for surgical aortic valve replacement, minimally invasive approaches have garnered interest during the past decade. Since the introduction of right anterior thoracotomy for aortic valve replacement in 1993, right anterior mini-thoracotomy and upper hemi-sternotomy have become the predominant incisional approaches among cardiac surgeons performing minimal access aortic valve replacement. Beside the location of the incision, the arterial cannulation site represents the second major landmark of minimal access techniques for aortic valve replacement. The two most frequently used arterial cannulation sites include central aortic and peripheral femoral approaches. With the purpose of reducing surgical trauma in these patients, we have opted for a right anterior mini-thoracotomy approach with a central aortic cannulation site. This protocol describes in detail a technique for minimally invasive aortic valve replacement and provides recommendations for patient selection criteria, including cardiac computer tomography measurements. The indications and limitations of this technique, as well as its alternatives, are discussed.
Among heart valve lesions diagnosed as hemodynamically relevant and clinically receiving particular attention, aortic valve stenosis is the most common valvular pathology in the United States and developed countries1,2. In the Cardiovascular Health Study, 2% of patients had frank aortic stenosis, with a clear increase in prevalence with growing age: 1.3% in patients aged 65-75 years, 2.4% in those aged 75-85 years, and 4% in patients older than 85 years1. For symptomatic patients presenting with severe aortic valve stenosis, aortic valve replacement is a Class I recommendation in the guidelines of the American Heart Association for the management of patients with valvular heart disease3.
Conventional surgical aortic valve replacement through median full sternotomy (FS) has been established as the gold standard for treating aortic valve stenosis with excellent results in terms of morbidity and mortality4. These results have encouraged the extension of therapeutic indications to older patients and patients with a higher risk profile. A number of treatment strategies have been implemented in these patient subsets to maintain the same good results achieved by conventional surgical aortic valve replacement in the general population. Among these alternative treatment modalities, transcatheter aortic valve implantation (TAVI) was introduced in 2002 by Cribier and colleagues5. Performed initially in moribund patients, TAVI has rapidly emerged as the treatment of choice for patients with severe aortic stenosis who are not suitable for conventional surgical aortic valve replacement6,7, or as a less invasive approach for surgery for patients at high risk8,9.
Despite the improved outcomes of TAVI in selected patient subsets, many patients with symptomatic aortic valve stenosis are still candidates for surgical aortic valve replacement. In these patients, FS aortic valve replacement is the most frequently used approach by cardiac surgeons. Nevertheless, various 'minimally invasive' techniques have been developed with the rationale of reducing surgical trauma10. All these minimal-access techniques have aimed at improving patient comfort by reducing post-operative pain and accelerating patient recovery by shortening the hospital stay and potentially saving global costs10. Among minimally invasive incisional approaches upper hemi-sternotomy (UHS) and right anterior mini-thoracotomy (RAMT) have become the predominant techniques reported in the literature11. Right anterior mini-thoracotomy for aortic valve replacement was initially reported by Benetti et al.12, and upper hemi-sternotomy was first described by several authors11. In addition to incisional alternatives, two arterial perfusion strategies are currently used: i) peripheral femoral arterial cannulation, which is more frequently employed than ii) central aortic cannulation.
In spite of reported improvement in patient outcomes following minimally invasive aortic valve replacement, concerns about the disadvantages of restricted operative field and peripheral arterial perfusion strategies13 lead many cardiac surgeons to not let their patients benefit from potential advantages of minimal access approaches for aortic valve replacement. The goal of this protocol is to describe in detail this technique of minimally invasive aortic valve replacement through a right anterior mini-thoracotomy without rib resection/fracture, and with central aortic cannulation for arterial perfusion. By following this protocol, a larger number of cardiac surgeons can perform right anterior mini-thoracotomy for aortic valve replacement in certain patient groups. Patient selection and limitations of the technique are discussed. Early results are compared to those of a cohort of patients undergoing isolated aortic valve replacement by full sternotomy.
The protocol follows our institutional guidelines of the human research ethics committee.
1. Patient Selection (Table 1)
2. Preparation for Surgery
3. Surgery
4. Post-Operative Patient Care
Statistical analysis is done for continuous variables (presented as means ± SD) in Table 2, Table 3, and Table 4 using the non-parametric Mann Whitney test. Categorical variables are presented as percentages in Table 2, Table 3, and Table 4, and are compared by the Chi-square test. The statistical analyses are performed using commercially available software, with a statistical significance threshold set at p<0.05.
Table 2 depicts patient characteristics. Aortic valve replacement was done according to the described protocol of right anterior mini-thoracotomy in 196 patients who fulfilled the selection criteria (Table 1). Patient characteristics were similar to 171 patients undergoing aortic valve replacement through full sternotomy during the same period. The adoption of the approach was dictated by the choice of the patient, and fulfillment of the selection criteria, including technical feasibility according to the CT measurements (Table 1).
Intra operative data are shown in Table 3. Based on the pre-operative planning CT, access to the pleural space and the heart was secured through the third intercostal space in 80%, and the second in 20%, of the patients. In one patient, after the start of the cardio-pulmonary bypass, increasing perimeter of the abdomen and volume loss was observed. With the suspicion of an abdominal problem, a laparotomy was performed and an iatrogenic tear in the iliac vein of the percutaneously cannulated side was repaired. In order to shorten the cardiac and global operative times, the incision was extended by a transverse sternotomy and the aortic valve replacement performed afterwards in standard manner. The patient did well post-operatively. The data from this patient are included in the right mini-thoracotomy group. As expected, compared to the full sternotomy approach, a right anterior mini-thoracotomy necessitated significantly longer ischemic, cardio-pulmonary, and operative times. The proportion of biological versus mechanical valve substitutes did not differ between the groups.
Table 4 demonstrates early post-operative outcomes. In spite of longer operation times, with the exception of a peripheral cannulation complication, right anterior mini-thoracotomy did not increase the rate of major adverse cardiovascular and cerebrovascular events in comparison to full sternotomy incision. Patients undergoing right anterior mini-thoracotomy tended to be extubated earlier and stayed for a shorter period in the intensive care unit than those operated through full sternotomy, although the differences were not statistically significant. However, compared to full sternotomy, right anterior mini-thoracotomy significantly reduced the need for pain medication and transfusion requirements, the incidence of new onset atrial fibrillation and deep wound infection as well as the global length of hospital stay. In addition, patients reported great satisfaction with the cosmetic results (Figure 5).
Figure 1:3-D chest CT reconstruction for pre-operative planning. Measurements from the mid-clavicular line to the aortic cannulation site (origin of the right innominate artery) and to the aortic valve of the second and third intercostal spaces (ICS) are reported on this image (1 mm in scale = 2.9 mm). In this case, the second ICS is chosen for incisional approach of the mini-thoracotomy while with comparable distances to the aortic valve, the distance to the cannulation site is shorter for the second compared to the third ICS. Please click here to view a larger version of this figure.
Figure 2: CT measurements for pre-operative planning from the same patient as in Figure 1. (A) Annulus sizing. The largest and shortest aortic annulus are both greater than the 20 mm criterium for selection of the patients given in Table 1. (B) Distance from the left coronary ostium to the aortic annulus. This distance is greater than 12 mm criterium for selection of the patients given in Table 1. (C) Distance from the right coronary ostium to the aortic annulus. This distance is greater than 12 mm criterium for selection of the patients given in Table 1 (1 mm in scale = 2.9 mm). CT measurements of this patient fulfills the CT criteria for selection of the patients for right anterior mini-thoracotomy with pleural entry in the second ICS. RV = right ventricle, LA = left atrium, dAo = descending Aorta, LMCA = left main coronary artery, LV = left ventricle, RCA = right coronary artery, Ao = ascending aorta. Please click here to view a larger version of this figure.
Figure 3: Bi-atrial view by transesophageal echocardiography. The venous cannula is inserted percutaneously through the right femoral vein and placed through the inferior vena cava (IVC) up to the origin of the superior vena cava (SVC) (1 mm in scale = 2.9 mm). LA = left atrium, RA = right atrium. Please click here to view a larger version of this figure.
Figure 4: Global view of the cardiopulmonary set-up. The head of the patient is to the left of the picture. Please click here to view a larger version of this figure.
Figure 5: Wound cosmetic. One week after aortic valve replacement through right anterior mini-thoracotomy, the general appearance of the wound is satisfactory. Please click here to view a larger version of this figure.
Absence of: | |
chest deformities | |
previous right hemi-thorax surgery or irradiation | |
aneurysm of the ascending aorta | |
emergency operation | |
active endocarditis | |
Optimal measurements calculated from chest CT | |
Intercostal space to aortic valve | < 120 mm |
Intercostal space to aortic cannulation site* | < 120 mm |
Aortic valve annulus diameter | > 20 mm |
Right coronary ostium to aortic valve annulus | > 12 mm |
Left coronary ostium to aortic valve annulus | > 12 mm |
* Origin of the right innominate artery |
Table 1: Selection criteria for right anterior mini-thoracotomy for isolated aortic valve replacement. After exclusion of patients with major chest deformities, previous right hemi-thorax irradiation or surgery and ascending aortic aneurysms, optimal measurements provided by chest CT allow the selection of the patient for the procedure and the choice of intercostal space for incision.
Right anterior mini-thoracotomy | Full sternotomy | p value | |
N | 196 | 171 | |
Age, years | 70±10 | 68±13 | 0.26 |
LVEF < 0.35 (%) | 2 (1) | 4 (2.3) | 0.2 |
Hematocrit % | 38.9±2 | 37.2±1.6 | 0.09 |
Native valve disease | 0.16 | ||
AS | 155 | 135 | |
AI | 34 | 27 | |
AS+AI | 7 | 9 | |
COPD (%) | 13 (6.6) | 14 (8.2) | 0.8 |
CVD (%) | 17 (8.7) | 16 (9.4) | 0.9 |
PVD (%) | 10 (5.1) | 11 (6.4) | 0.14 |
Diabetes mellitus (%) | 29 (14.8) | 18 (10.5) | 0.42 |
Arterial hypertension (%) | 91 (46) | 105 (61) | 0.27 |
CHF (%) | 3 (1.5) | 8 (4.7) | 0.26 |
Euroscore* | 3.8±2.0 | 2.6±2.4 | 0.7 |
LVEF: left ventricular ejection fraction | |||
AS: aortic stenosis | |||
AI: aortic insufficiency | |||
COPD: chronic obstructive pulmonary disease | |||
CVD: cerebro-vascular disease | |||
PVD: peripheral vascular disease | |||
CHF: congestive heart failure | |||
* Reference 15 |
Table 2: Patients' characteristics. The pre-operative condition of the patients undergoing right anterior mini-thoracotomy is comparable to that of patients operated by a full sternotomy approach.
Right anterior mini-thoracotomy | Full sternotomy | p value | |
N | 196 | 171 | |
Operative time (min) | 274±60 | 209±48 | 0.0001 |
CPB time (min) | 157±36 | 110±27 | 0.0001 |
Cross-clamp time (min) | 108±23 | 62±15 | 0.0001 |
Biological valve (%) | 161 (82) | 121 (71) | 0.09 |
CPB: cardiopulmonary bypass |
Table 3: Intra-operative data. Operation times and type of the valve substitute of the two cohorts of right anterior mini-thoracotomy and full sternotomy approaches.
Right anterior mini-thoracotomy | Full sternotomy | p value | |
N | 196 | 171 | |
Early mortality (%) | 3 (1.5) | 3 (1.8) | 0.1 |
Re-op. for bleeding (%) | 4 (2) | 4 (2.3) | 0.6 |
Peri-op. MI (%) | 1 (0.5) | 2 (1.2) | 0.3 |
Stroke (%) | 5 (2.6) | 6 (3.5) | 0.7 |
Ventilation time (h) | 10.2±6.3 | 14.4±20 | 0.09 |
ICU stay (d) | 2±3.3 | 3.7±14 | 0.4 |
New onset Afib (%) | 35 (18) | 51 (30) | 0.02 |
RBC’s (pack/patient) | 1.6±2.8 | 2.1±2.1 | 0.002 |
MS (mg/day)* | 39.7±22.8 | 52.5±26.3 | 0.007 |
Deep wound infection (%) | 0 (0) | 2 (1.2) | 0.02 |
Hospital stay (d) | 8.2±2.7 | 10.2±7.3 | 0.04 |
MI: myocardial infarction | |||
ICU: intensive care unit | |||
Afib: atrial fibrillation | |||
RBC: Red blood cells | |||
MS: Morphine sulfate for the first 5 post-operative days | |||
* for the first 3 days |
Table 4: Early outcomes. In spite of longer operation times, right anterior mini-thoracotomy resulted in better early outcomes without increasing major adverse events.
In this protocol, we describe in detail the technique of right anterior mini-thoracotomy for isolated aortic valve replacement, and highlight the patient selection criteria for this procedure. As for any other therapeutic intervention, proper patient selection is the key to successful accomplishment of the procedure. The optimal CT measurements for consideration of patients for this technique are precisely described in this protocol, and are based on experience and consider the extensive work of Dr. Glauber and coworkers in this field10. These optimal CT measurements add some valuable criteria, i.e., the annulus size, distances of the coronary ostia from the aortic annulus, to the recommendations of Glauber et al., who find patients suitable for this technique only if: (1) the ascending aorta is rightward (more than one half located on the right in respect to the right sternal border) at the level of the main pulmonary artery, (2) the distance between the ascending aorta and the sternum does not exceed 10 cm, and (3) the angle between the midline and the inclination of ascending aorta (α angle) is larger than 45°10. These CT selection criteria differ somewhat from those recommended by Glauber et al.10. First, the 3-D CT reconstruction of the distance from the mid-clavicular line of the second and third intercostal spaces to the origin of the right innominate artery is of utmost importance in this protocol. The shorter this distance is, the safer the central arterial cannulation. Second, beside the configuration of the chest 3-D CT measurements of the distance from the mid-clavicular line of the second and third intercostal spaces to the aortic valve annulus are mandatory for estimation of the ease of the valve procedure itself. Thus, selection of patients for this protocol is based on two key distances, i.e. working distances to the aortic valve and the central arterial cannulation site. As a consequence, the criteria of distance from the intercostal space to the aortic annulus and the aortic cannulation site give more precise information and are more helpful for selection and pre-operative planning of patients than the formulation of distance of ascending aorta to the sternum used by Glauber and co-workers10. Other groups do not routinely use CT measurements for selection of patients for right anterior mini-thoracotomy16,17,18. However, in order to compensate for skipping this step, these authors manage a dislocation of the ribs during entry to the pleural space followed by refixation of the dislocated rib(s) with a plate at the end of the operation16,17,18. However, this strategy may reduce patient comfort and could potentially lead to more bleeding from broken bone surfaces.
An important aspect of the present protocol is that the arterial perfusion is done through central aortic cannulation. The site of arterial perfusion during minimal access heart valve surgery in general, and during right anterior mini-thoracotomy in particular, remains controversial. While a number of centers and surgeons have reported an increased risk of stroke following peripheral femoral arterial cannulation as compared to central aortic cannulation19,20, other authors have reported an increased risk of stroke following central aortic cannulation21, or no difference is reported between the two strategies22. This disparity might be due not only to different definitions of stroke, but also to associated comorbidities such as older age and peripheral arterial vascular disease20,23. Other potential complications related to peripheral femoral arterial perfusion are infection, lymphoid fistula, arterial wall dissection, and distal limb ischemia24. In contrast to Glauber and colleagues, who used short-tipped aortic cannulas for central aortic cannulation25, we used commercially available arterial cannulas long enough so that the tip of the cannula could be placed at the origin of the descending aorta. The rationale of this strategy is to avoid the jet of the arterial perfusion into the aortic arch with potential dislocation and embolization of the arch vessels.
In this protocol, we gain access to the pleural space and the heart without deliberate fracture or dislocation of the ribs, in contrast to other protocols16,17,18. Based on copious experience, a gentle and gradual opening of the spreader avoids inadvertent fracture of the ribs. In the same line of tissue preservation, we do not systematically ligate the right internal thoracic vessels and reserve the ligation for anatomical variations of these vessels where they are laterally more than 1 cm to the right of the sternal edge. Systematic fracture of the ribs could potentially increase the amount of bleeding from the fractured bone surfaces and cause patients discomfort, despite the fixation at the end of the operation25.
Early mortality of patients operated on according to the present protocol of right anterior mini-thoracotomy (1.5%) compares favorably with those undergoing full sternotomy (1.8%) (Table 4) and with the reported outcomes for isolated aortic valve replacement in the STS Database (2.4%)26. Compared to full sternotomy, right mini-thoracotomy did not increase early postoperative complications, including reoperation for bleeding and stroke. The rate of reoperation in the two groups of patients (RAMT 2%, FS 2.3%) was slightly less than that reported in the STS database (3.9%), whereas the incidence of stroke was slightly higher (RAMT 2.6%, FS 3.5%, STS 1.1%)26. The reason for this difference might lie in the definition of stroke used in these patients and in the STS Database. In this protocol, all clinically diagnosed peri-operative cerebrovascular events, including reversible transient deficits were computed as stroke, while the STS Database reported on permanent stroke26. Further, in the studied patients, the incidence of new onset atrial fibrillation was reduced by RAMT approach (18%) compared to the FS group (30%, p=0.02) and the STS Database (31%)26. This finding is in accordance with the study of Glauber et al. reporting 18.1% and 27.9% of new onset atrial fibrillation in RAMT versus FS patients (p = 0.03)27. This difference might be due to the elimination of direct cannulation of the right atrium in the RAMT compared to the FS approach. In fact, the scar of the cannulation site of the right atrium may be the local origin of atrial arrhythmias. This hypothesis warrants further investigation. Another improvement in early patient outcomes achieved by the RAMT (0%) compared to FS approach (1.2%, p=0.02) was the disappearance of deep wound infection. In comparison to FS, RAMT access significantly lowered the need for pain medication and transfusion requirements in this protocol. As a result of reducing the incidence of post-operative new onset atrial fibrillation, transfusion requirements, and deep wound infection, as well as of enhancing patient comfort by reducing pain medication, the RAMT approach significantly shortened the length of hospital stay as compared to the FS approach. This observation is in accordance with other studies reporting similar reduction in the length of hospital stay, which could potentially contribute to cost savings16,27.
In this protocol, excision of the aortic valve and implantation of the new valve prosthesis remain unchanged compared to the full sternotomy. This may explain the achievement of the same good results of conventional AVR through full sternotomy, and makes this approach an attractive alternative to TAVI in the intermediate risk subgroup of patients28.
The limitations of the comparisons between the RAMT and FS cohorts in this protocol are due to the retrospective nature of these data. Although the risk profiles of the two groups reflected by Euroscore15 were comparable between these groups, other uncontrolled parameters might have affected the reported comparison. The limitations of the reported technique in this protocol are represented by the optimal CT measurements and other criteria reported in Table 1. In these patients, aortic valve replacement should be performed by conventional full or partial sternotomy access.
In conclusion, right anterior mini-thoracotomy as described in this protocol can be done in selected patients according to the described criteria with comparably good results achieved with the full sternotomy approach. The additional operative times required are not detrimental to the patients and are rewarded by improved patient comfort and accelerated early recovery. This minimally invasive approach can be a viable alternative to TAVI in intermediate risk patient populations needing an aortic valve replacement.
The authors have nothing to disclose.
This work was supported by a grant (N° 32119) of the Swiss Cardiovascular Foundation to RT.
Heart surgery infrastructure: | |||
Heart Lung Machine | Stockert | SIII | |
EOPA 24Fr. arterial cannula | Medtronic | 77624 | |
FemFlex arterial cannula | Edwards | FEMII20A | |
Quickdraw 25Fr. femoral venous cannula | Edwards | QD25 | |
Biomedicus 25Fr. Nextgen venous cannula | Medtronic | 96670-125 | |
LV vent catheter 17Fr. | Edwards | E061 | |
Antegrade 9Fr. cardioplegia cannula | Edwards | AR012V | |
Coronary artery ostial cannula 90° | Medtronic | 30155 | |
Coronary artery ostial cannula 45° | Medtronic | 30255 | |
Soft tissue retractor | |||
STAR soft tissue atraumatic retractor | Estech | EC400220 | |
Soft tissue retractor | Edwards | TRM | |
Electrocautery | Covidien | Force FXTM | |
Sutures: | |||
Polypropylene 4/0 | Ethicon | 8871H | |
Polypropylene 5/0 | Ethicon | 8870H | |
Braided polyesther 2/0 ligature with polybutylate coating | Ethicon | X305H | |
Braided polyesther2/0 with pledgets V5 | Ethicon | MEH7715N | |
Braided polyglactin 2/0 suture | Ethicon | V114H | |
Braided polyglactin 0 suture | Ethicon | W9996 | |
Drugs: | |||
Midazolam | Roche Pharma | N05CD08 | |
Rocuronium | MSD Merck Sharp & Dohme | M03AC09 | |
Propofol | Fresenius Kabi | N01AX10 | |
Fentanil | Actavis | N01AH01 | |
Heparin | Braun | B01AB01 | |
Protamin | MEDA Pharmaceutical | V03AB14 | |
Custodiol cardioplegia solution | Dr. F. Köhler Chemie GmbH | B05CX10 | |
Instruments: | |||
Window access retractor SI | Estech | 400-400 | |
SI retractor blade 40W50L | Estech | 400-172 | |
Ceramo atraumatic forceps 2.8×15/350 | Fehling | FE-MRA-3 | |
Ceramo HCR valve forceps 3.0×15/350 | Fehling | FE-MRA-0 | |
Ceramo HCR needle holder 2×10/340 | Fehling | FE-MRB-2 | |
Ceramo TC HCR needle holder curved 3×10/340 | Fehling | FE-MRG-9 | |
Ceramo HCR valve scissors 350 | Fehling | FE-MRA-7 | |
Ceramo HCR curved scissors 350 | Fehling | FE-MRA-6 | |
Cygnet flexible arched aortic clamp | Vitalitec | V10143 | |
Intrack insert set double traction | Vitalitec | N10122 | |
Dissection forceps Carpentier | Delacroix-Chevalier | DC13110-28 | |
Scissors Metzenbaum | Delacroix-Chevalier | B351751 | |
Needle holder Ryder | Delacroix-Chevalier | DC51130-20 | |
Dissection forceps DeBakey | Delacroix-Chevalier | DC12000-21 | |
Lung retractor | Delacroix-Chevalier | B803990 | |
Allis clamp | Delacroix-Chevalier | DC45907-25 | |
O’Shaugnessy Dissector | Delacroix-Chevalier | B60650 | |
18 blade knife | Delacroix-Chevalier | B130180 | |
11 blade knife | Premiere | 9311-2PK | |
Leriche haemostatic clamp | Delacroix-Chevalier | B86555 | |
Data analysis | |||
Mann-Whitney and Chi-square tests | GraphPad | Prism 7 |