© 2014 Wiley Periodicals, Inc.
343
SURGICAL TECHNIQUE ___________________________________________________________
Minimally Invasive Mitral Valve Surgery Utilizing Heart Port Technology Prashanth Vallabhajosyula, M.D., M.S., Tyler J. Wallen, D.O., Lauren P. Solometo, B.S., Jeanne Fox, C.R.N.P., William J. Vernick, M.D., and W. Clark Hargrove III, M.D. Department of Cardiac Surgery, University of Pennsylvania Health System, Philadelphia, Pennsylvania ABSTRACT Objective: To determine operative outcomes of right mini-thoracotomy mitral valve surgery utilizing port access technology in first-time and reoperative cardiac surgery patients. Methods: From 2002 to 2011, 881 patients underwent minimally invasive mitral valve surgery. Of these, 154 patients had previous cardiac operations via sternotomy (Group 1), of which 18 (12%) had two previous operations. Seven hundred and twenty-seven patients had no previous cardiac operations (Group 2). Results: Patient demographics were similar in both groups. In Group 1, 76 (49%) patients had previous coronary artery bypass grafting, 13 (8%) had previous aortic valve surgery, and 57 (37%) had previous mitral valve surgery. Preoperative echo findings for Groups 1 and 2 included severe mitral regurgitation (MR) (88%, n = 135; 94%, n = 687), mitral stenosis (MS) (4%, n = 6; 2%, n = 12), MS R MR (8%, n = 13; 4%, n = 28), and ejection fraction (48%, 56%). Operative procedures in Groups 1 and 2 were MV repair (54%, n = 84; 89%, n = 645) and MV replacement (46%, n = 70; 11%, n = 82). Circulatory management techniques for Groups 1 and 2 included endoballoon (75%, n = 116; 79%, n = 576), Chitwood clamp (8%, n = 12; 20%, n = 147), and fibrillatory arrest (17%, n = 30; 0.5%, n = 4). Perioperative outcomes were: stroke: 2.5%, 1.6%; reoperation for bleeding: 5%, 6%; valvular reoperation rate: 0.6%, 2%; aortic dissection: 2.5%, 1%; and wound infection: 0%, 0%. Transfusion requirement was 49% (n = 76) and 31% (n = 232), respectively. Median hospital stay was seven and seven days, respectively. On postoperative echocardiography, 98% (n = 151) and 99% (n = 718) of patients had zero or trace MR (1R) with 100% freedom from MR > 2R. In-hospital mortality was 3% (n = 5) and 1% (n = 8). Conclusions: Operative outcomes with minimally invasive mitral valve surgery utilizing port access technology can be performed safely. Stroke rate was higher in the reoperative cases (p = NS) although similar to reports evaluating redo sternotomy in mitral valve cases. doi: 10.1111/jocs.12293 (J Card Surg
2014;29:343–348) Several studies have reported good outcomes with minimally invasive mitral valve surgery utilizing different cannulation and aortic occlusion strategies.1–4 In a recent analysis of the Society of Thoracic Surgeons (STS) database, minimally invasive mitral valve approach was associated with lower transfusion requirement, lower hospital length of stay, and similar operative mortality to conventional sternotomy approach in firsttime cases.1 In first-time cases, minimally invasive surgery was associated with longer cardiopulmonary Conflict of interest: The authors acknowledge no conflict of interest in the submission. Address for correspondence: Tyler J. Wallen, D.O., University of Pennsylvania Medical Center, 3400 Spruce St, 6 Silverstein, Philadelphia, PA 19104. Fax: 2153495798; e-mail: pvallabhajosyula@uphs. upenn.edu
and aortic cross-clamp times, and most importantly, with significantly higher stroke rate. Overall, the study attested to the validity of employing minimally invasive techniques in first-time mitral valve surgery. The study excluded patients who had concomitant valvular procedures, coronary artery bypass grafting (CABG), or any other cardiac operations. Also, all reoperative cases were excluded. The study also did not evaluate echocardiographic preoperative and postoperative mitral valve function in the minimally invasive versus conventional sternotomy approaches. At our institution, minimally invasive mitral valve surgery is performed utilizing endoballoon aortic occlusion as the preferred technique, with direct aortic occlusion using the Chitwood clamp (Scanlan International, St. Paul, MN, USA) as the alternative strategy. The minimally invasive approach has also
344
VALLABHAJOSYULA, ET AL. REDO MININALLY INVASIVE MITRAL VALVE SURGERY
been our preferred technique in reoperative mitral valve surgery. The utility of minimally invasive mitral valve surgery in reoperative cardiac surgery has been previously described.2–5 In these cases, operative mortality was low, with good postoperative outcomes, although mitral valve function was not evaluated in many studies. Other studies have shown that reoperative mitral valve surgery is associated with higher morbidity, especially a higher stroke rate, than open sternotomy.6 In this study, we report a single-institution experience with minimally invasive mitral valve surgery, with a specific focus on assessing operative outcomes in redo mitral valve surgery. METHODS Institutional Review Board approval was obtained to retrospectively review the medical records of all patients who underwent minimally invasive mitral valve surgery at our institution. Over 90% of the mitral valve surgeries performed over the last 10 years have been via the right minithoracotomy approach. The median sternotomy approach, including in reoperative cases, was primarily pursued only in situations where patients needed concomitant cardiac operations not feasible via the right thoracotomy approach or femoral/axillary artery cannulation was not possible (Fig. 1).
Figure 1. Surgical volume and patient assignment by operative approach and incidence. CABG, coronary artery bypass grafting.
J CARD SURG 2014;29:343–348
Figure 2. Aortic occlusion techniques.
In performing mitral valve surgery with a right minithoracotomy approach, two aortic occlusion techniques were employed: endoballoon occlusion, the Chitwood cross-clamp technique, and fibrillatory arrest when aortic occlusion was deemed unsafe. Our preferred aortic occlusion technique has been the Heart Port endoballoon occlusion technique via the femoral artery (79%). In situations where the endoballoon technique was not possible due to aorto-iliac disease or severe arch disease, our second preference has been direct aortic cannulation and placement of the Chitwood cross-clamp. In situations where manipulation of the aorta is prohibitive, fibrillatory arrest was utilized to perform mitral valve operations, although we tend to avoid this technique as much as possible (Fig. 2). From January 2002 to December 2011, 881 consecutive patients (494 males, 387 females), mean age of 59 years, underwent right mini-thoracotomy mitral valve surgery. We defined minimally invasive mitral valve surgery as open-heart surgery conducted through a mini right anterolateral thoracotomy (4–5 cm) utilizing Heart Port technology (Cardiovations, Ethicon, Inc, Somerville, NJ, USA). The technique has been previously described.7 All patients receive a preoperative trans-esophageal echocardiogram (TEE) and the distal thoracic is screened for atherosclerotic disease. If present the patient receives a CT angiogram of the chest/abdomen/pelvis. Under general anesthesia with endotracheal intubation, a TEE probe, a Swan-Ganz catheter, and a superior vena cava cannula (16 French) were placed, along with bilateral radial arterial lines. A right anterolateral thoracotomy was performed (4–5 cm), typically in the 4th intercostal space. Femoral cutdown was performed via a 2- to 3-cm incision, and arterial cannulation was achieved via the femoral artery with a 21 French Y-cannula (Edwards Lifesciences, Irvine, CA, USA). In cases where femoral arterial access was prohibitive, axillary cannulation was performed, again with a 21 French Y-cannula. Venous cannulation was obtained via a 25 French venous cannula placed in the femoral vein. Patients were placed on cardiopulmonary bypass and cooled to 28–328C. Aortic occlusion was typically achieved using endoballoon under echocardiography guidance. In a small percentage of cases, the Chitwood (18%, 159/881) aortic cross-clamp and fibrillatory arrest (3%, 30/881) were used with mean aortic pressures maintained at or above 60 mmHg. The left atrium is opened at the atrial groove. When endoballoon aortic occlusion is not feasible, the Chitwood clamp technique was the preferred alternative. If this was not possible, fibrillatory arrest was employed. Antegrade crystalloid
345
J CARD SURG 2014;29:343–348
VALLABHAJOSYULA, ET AL. REDO MININALLY INVASIVE MITRAL VALVE SURGERY
cardioplegia was used in all cases. Carbon dioxide is continuously insufflated into the chest throughout the procedure. At the conclusion of the procedure the patient is re-warmed and weaned from cardiopulmonary bypass and closed in the standard fashion.
approach were performed. Of these, 881 cases had mitral valve surgical intervention. One hundred and fiftyfour patients had previous cardiac operations and 727 were first-time procedures. The breakdown of previous operations included the following: CABG (49%), previous aortic valve procedure (8%), and previous mitral valve surgery (37%). Of note, more than one-third of the redo cases had previous mitral valve surgery via median sternotomy approach. Thirteen percent of the reoperative cases were second time redo operations (Fig. 1).
Definitions Hospital mortality: death for any reason occurring within 30 days after the surgery or during the same hospitalization. Major neurological event/stroke: evidence in the postoperative period of a new central neurological defect that did not resolve by discharge. Valvular reoperation: reoperation within 30 days of the original procedure for valvular dysfunction, including intraoperative reoperation on the mitral valve. Reoperation for bleeding: reexploration of the chest during the postoperative period for suspected ongoing bleeding during the same hospitalization. Type-A dissection: newly present aortic dissection confirmed by TEE in the operating room. Non-redo cases were placed on a postoperative ‘‘fast track’’ protocol with a goal of extubation within four hours of arrival in the intensive care unit. Reoperative cases were managed expectantly.
Patient characteristics The average patient in the reoperative group was older, had a higher incidence of New York Heart Classification > 2, had higher rates of diabetes, hypertension, atrial fibrillation, and coronary artery disease (p < 0.05). The left ventricular ejection fraction (LVEF) was also lower in the reoperative cohort as compared to the first-time patients (p < 0.05). Thirteen percent of cases in the reoperative group were second time redo cases. Overall, in comparing the two groups, the reoperative patients had greater comorbidity, undergoing more complex operations (Table 1). Mitral valve pathology and operative outcomes
Data analysis Data are presented as mean standard deviation. Statistical analysis was completed using SPSS software (IBM, Inc., Armonk, NY, USA). A Fisher’s exact test was used to analyze categorical data and an unpaired t-test was used when analyzing continuous data. A p-value of 2 1st redo 2nd redo Diabetes Atrial fibrillation Hypertension Coronary artery disease LV EF (%) Mitral valve disease Stenosis Regurgitation Mixed
Reoperative (n = 154)
First Time (n = 727)
P
66 13 93 (60%) 102 (66%) 136 (88%) 18 (12%) 23 (15%) 64 (42%) 84 (55%) 82 (53%) 48 14
57 13 401 (55%) 210 (29%) N/A N/A 44 (6%) 158 (22%) 268 (37%) 40 (5%) 56 11
343
SURGICAL TECHNIQUE ___________________________________________________________
Minimally Invasive Mitral Valve Surgery Utilizing Heart Port Technology Prashanth Vallabhajosyula, M.D., M.S., Tyler J. Wallen, D.O., Lauren P. Solometo, B.S., Jeanne Fox, C.R.N.P., William J. Vernick, M.D., and W. Clark Hargrove III, M.D. Department of Cardiac Surgery, University of Pennsylvania Health System, Philadelphia, Pennsylvania ABSTRACT Objective: To determine operative outcomes of right mini-thoracotomy mitral valve surgery utilizing port access technology in first-time and reoperative cardiac surgery patients. Methods: From 2002 to 2011, 881 patients underwent minimally invasive mitral valve surgery. Of these, 154 patients had previous cardiac operations via sternotomy (Group 1), of which 18 (12%) had two previous operations. Seven hundred and twenty-seven patients had no previous cardiac operations (Group 2). Results: Patient demographics were similar in both groups. In Group 1, 76 (49%) patients had previous coronary artery bypass grafting, 13 (8%) had previous aortic valve surgery, and 57 (37%) had previous mitral valve surgery. Preoperative echo findings for Groups 1 and 2 included severe mitral regurgitation (MR) (88%, n = 135; 94%, n = 687), mitral stenosis (MS) (4%, n = 6; 2%, n = 12), MS R MR (8%, n = 13; 4%, n = 28), and ejection fraction (48%, 56%). Operative procedures in Groups 1 and 2 were MV repair (54%, n = 84; 89%, n = 645) and MV replacement (46%, n = 70; 11%, n = 82). Circulatory management techniques for Groups 1 and 2 included endoballoon (75%, n = 116; 79%, n = 576), Chitwood clamp (8%, n = 12; 20%, n = 147), and fibrillatory arrest (17%, n = 30; 0.5%, n = 4). Perioperative outcomes were: stroke: 2.5%, 1.6%; reoperation for bleeding: 5%, 6%; valvular reoperation rate: 0.6%, 2%; aortic dissection: 2.5%, 1%; and wound infection: 0%, 0%. Transfusion requirement was 49% (n = 76) and 31% (n = 232), respectively. Median hospital stay was seven and seven days, respectively. On postoperative echocardiography, 98% (n = 151) and 99% (n = 718) of patients had zero or trace MR (1R) with 100% freedom from MR > 2R. In-hospital mortality was 3% (n = 5) and 1% (n = 8). Conclusions: Operative outcomes with minimally invasive mitral valve surgery utilizing port access technology can be performed safely. Stroke rate was higher in the reoperative cases (p = NS) although similar to reports evaluating redo sternotomy in mitral valve cases. doi: 10.1111/jocs.12293 (J Card Surg
2014;29:343–348) Several studies have reported good outcomes with minimally invasive mitral valve surgery utilizing different cannulation and aortic occlusion strategies.1–4 In a recent analysis of the Society of Thoracic Surgeons (STS) database, minimally invasive mitral valve approach was associated with lower transfusion requirement, lower hospital length of stay, and similar operative mortality to conventional sternotomy approach in firsttime cases.1 In first-time cases, minimally invasive surgery was associated with longer cardiopulmonary Conflict of interest: The authors acknowledge no conflict of interest in the submission. Address for correspondence: Tyler J. Wallen, D.O., University of Pennsylvania Medical Center, 3400 Spruce St, 6 Silverstein, Philadelphia, PA 19104. Fax: 2153495798; e-mail: pvallabhajosyula@uphs. upenn.edu
and aortic cross-clamp times, and most importantly, with significantly higher stroke rate. Overall, the study attested to the validity of employing minimally invasive techniques in first-time mitral valve surgery. The study excluded patients who had concomitant valvular procedures, coronary artery bypass grafting (CABG), or any other cardiac operations. Also, all reoperative cases were excluded. The study also did not evaluate echocardiographic preoperative and postoperative mitral valve function in the minimally invasive versus conventional sternotomy approaches. At our institution, minimally invasive mitral valve surgery is performed utilizing endoballoon aortic occlusion as the preferred technique, with direct aortic occlusion using the Chitwood clamp (Scanlan International, St. Paul, MN, USA) as the alternative strategy. The minimally invasive approach has also
344
VALLABHAJOSYULA, ET AL. REDO MININALLY INVASIVE MITRAL VALVE SURGERY
been our preferred technique in reoperative mitral valve surgery. The utility of minimally invasive mitral valve surgery in reoperative cardiac surgery has been previously described.2–5 In these cases, operative mortality was low, with good postoperative outcomes, although mitral valve function was not evaluated in many studies. Other studies have shown that reoperative mitral valve surgery is associated with higher morbidity, especially a higher stroke rate, than open sternotomy.6 In this study, we report a single-institution experience with minimally invasive mitral valve surgery, with a specific focus on assessing operative outcomes in redo mitral valve surgery. METHODS Institutional Review Board approval was obtained to retrospectively review the medical records of all patients who underwent minimally invasive mitral valve surgery at our institution. Over 90% of the mitral valve surgeries performed over the last 10 years have been via the right minithoracotomy approach. The median sternotomy approach, including in reoperative cases, was primarily pursued only in situations where patients needed concomitant cardiac operations not feasible via the right thoracotomy approach or femoral/axillary artery cannulation was not possible (Fig. 1).
Figure 1. Surgical volume and patient assignment by operative approach and incidence. CABG, coronary artery bypass grafting.
J CARD SURG 2014;29:343–348
Figure 2. Aortic occlusion techniques.
In performing mitral valve surgery with a right minithoracotomy approach, two aortic occlusion techniques were employed: endoballoon occlusion, the Chitwood cross-clamp technique, and fibrillatory arrest when aortic occlusion was deemed unsafe. Our preferred aortic occlusion technique has been the Heart Port endoballoon occlusion technique via the femoral artery (79%). In situations where the endoballoon technique was not possible due to aorto-iliac disease or severe arch disease, our second preference has been direct aortic cannulation and placement of the Chitwood cross-clamp. In situations where manipulation of the aorta is prohibitive, fibrillatory arrest was utilized to perform mitral valve operations, although we tend to avoid this technique as much as possible (Fig. 2). From January 2002 to December 2011, 881 consecutive patients (494 males, 387 females), mean age of 59 years, underwent right mini-thoracotomy mitral valve surgery. We defined minimally invasive mitral valve surgery as open-heart surgery conducted through a mini right anterolateral thoracotomy (4–5 cm) utilizing Heart Port technology (Cardiovations, Ethicon, Inc, Somerville, NJ, USA). The technique has been previously described.7 All patients receive a preoperative trans-esophageal echocardiogram (TEE) and the distal thoracic is screened for atherosclerotic disease. If present the patient receives a CT angiogram of the chest/abdomen/pelvis. Under general anesthesia with endotracheal intubation, a TEE probe, a Swan-Ganz catheter, and a superior vena cava cannula (16 French) were placed, along with bilateral radial arterial lines. A right anterolateral thoracotomy was performed (4–5 cm), typically in the 4th intercostal space. Femoral cutdown was performed via a 2- to 3-cm incision, and arterial cannulation was achieved via the femoral artery with a 21 French Y-cannula (Edwards Lifesciences, Irvine, CA, USA). In cases where femoral arterial access was prohibitive, axillary cannulation was performed, again with a 21 French Y-cannula. Venous cannulation was obtained via a 25 French venous cannula placed in the femoral vein. Patients were placed on cardiopulmonary bypass and cooled to 28–328C. Aortic occlusion was typically achieved using endoballoon under echocardiography guidance. In a small percentage of cases, the Chitwood (18%, 159/881) aortic cross-clamp and fibrillatory arrest (3%, 30/881) were used with mean aortic pressures maintained at or above 60 mmHg. The left atrium is opened at the atrial groove. When endoballoon aortic occlusion is not feasible, the Chitwood clamp technique was the preferred alternative. If this was not possible, fibrillatory arrest was employed. Antegrade crystalloid
345
J CARD SURG 2014;29:343–348
VALLABHAJOSYULA, ET AL. REDO MININALLY INVASIVE MITRAL VALVE SURGERY
cardioplegia was used in all cases. Carbon dioxide is continuously insufflated into the chest throughout the procedure. At the conclusion of the procedure the patient is re-warmed and weaned from cardiopulmonary bypass and closed in the standard fashion.
approach were performed. Of these, 881 cases had mitral valve surgical intervention. One hundred and fiftyfour patients had previous cardiac operations and 727 were first-time procedures. The breakdown of previous operations included the following: CABG (49%), previous aortic valve procedure (8%), and previous mitral valve surgery (37%). Of note, more than one-third of the redo cases had previous mitral valve surgery via median sternotomy approach. Thirteen percent of the reoperative cases were second time redo operations (Fig. 1).
Definitions Hospital mortality: death for any reason occurring within 30 days after the surgery or during the same hospitalization. Major neurological event/stroke: evidence in the postoperative period of a new central neurological defect that did not resolve by discharge. Valvular reoperation: reoperation within 30 days of the original procedure for valvular dysfunction, including intraoperative reoperation on the mitral valve. Reoperation for bleeding: reexploration of the chest during the postoperative period for suspected ongoing bleeding during the same hospitalization. Type-A dissection: newly present aortic dissection confirmed by TEE in the operating room. Non-redo cases were placed on a postoperative ‘‘fast track’’ protocol with a goal of extubation within four hours of arrival in the intensive care unit. Reoperative cases were managed expectantly.
Patient characteristics The average patient in the reoperative group was older, had a higher incidence of New York Heart Classification > 2, had higher rates of diabetes, hypertension, atrial fibrillation, and coronary artery disease (p < 0.05). The left ventricular ejection fraction (LVEF) was also lower in the reoperative cohort as compared to the first-time patients (p < 0.05). Thirteen percent of cases in the reoperative group were second time redo cases. Overall, in comparing the two groups, the reoperative patients had greater comorbidity, undergoing more complex operations (Table 1). Mitral valve pathology and operative outcomes
Data analysis Data are presented as mean standard deviation. Statistical analysis was completed using SPSS software (IBM, Inc., Armonk, NY, USA). A Fisher’s exact test was used to analyze categorical data and an unpaired t-test was used when analyzing continuous data. A p-value of 2 1st redo 2nd redo Diabetes Atrial fibrillation Hypertension Coronary artery disease LV EF (%) Mitral valve disease Stenosis Regurgitation Mixed
Reoperative (n = 154)
First Time (n = 727)
P
66 13 93 (60%) 102 (66%) 136 (88%) 18 (12%) 23 (15%) 64 (42%) 84 (55%) 82 (53%) 48 14
57 13 401 (55%) 210 (29%) N/A N/A 44 (6%) 158 (22%) 268 (37%) 40 (5%) 56 11
