As we enter a new era of coronary artery revascularization, the treatment options for coronary artery disease remain in a state of constant flux, with endless advancements in minimally invasive coronary surgery. Robotics-assisted endoscopic single vessel small thoracotomy (endo-SVST) has been shown to be effective, feasible, and reproducible.1–4 It has resulted in reduction in surgical morbidity and length of stay in hospital and in return to functional health.
Endo-SVST is a procedure in evolution, requiring a continual improvement in techniques as new technologies become available. At present, the da Vinci Surgical System (Intuitive Surgical, Inc., Sunnyvale, CA) is the only robotic minimally invasive surgery system that has been approved by the Food and Drug Administration available for clinical use. Robotic assistance has resulted in decreased morbidity, while providing anastomotic and early graft quality comparable with conventional left internal thoracic artery (LITA) harvesting through sternotomy and anastomosis under cardioplegic arrest.5,6
The robotics-assisted procedure involves grafting the LITA to the left anterior descending (LAD) coronary artery on the beating heart using a stabilizing device around the target site.5 This allows the surgeon to perform the bypass anastomosis without the need for cardiopulmonary bypass (CPB).7 Robotic assistance has further reduced the invasiveness of the procedure by allowing the LITA to be harvested endoscopically through 1-cm incisions, while the anastomosis is performed through a 3- to 4-cm nonrib-spreading incision directly over the target coronary anastomosis site, providing better exposure and allowing for a much smaller incision.5,8 The main advantages of this approach are the avoidance of full sternotomy, CPB, and aortic manipulation that are a part of conventional coronary artery bypass grafting procedures.9 Potential beneficial patient outcomes include a reduction in morbidity, pain, and length of hospital stay without compromising the quality of the surgical procedure.10
Although the technology of remotely manipulated surgical instruments continues to advance, there are still some significant limitations that require further investigation and development. Current techniques for selecting patients for endo-SVST procedures are based on external anatomic features and information extracted from two-dimensional chest radiographs (Fig. 1). These methods of planning can lead to difficulties and errors in the operating room. In one study, an 86-patient evaluation of robotic coronary surgery, an erroneous anastomosis of the LITA to the coronary vein was performed.11 Another study of 22 patients undergoing totally endoscopic coronary artery bypass grafting with CPB observed that the LAD artery in 1 patient could not be identified and that the large diagonal branch was grafted instead of the LAD artery in another patient.12 The need for port repositioning or conversion to open surgery is commonly seen, especially in totally endoscopic procedures.13,14
In an attempt to solve some of these problems, animal studies have shown that the use of epicardial ultrasound in beating heart surgery is effective in detecting the location of the LAD artery.15 Other researchers have shown some benefit in the use of magnetic resonance coronary angiography to plan totally endoscopic procedures.16 However, a method for planning robotics-assisted cardiac procedures that not only determines the ideal location of the incisions but also assesses which patients are eligible for a minimally invasive approach, is needed. Ideally, a useful eligibility measure would be based on simple measurements performed on preoperative images; however, previous studies have not been able to establish a relationship between the ease of performance and the contour of the thoracic cavity (barrel and flat).12
The objectives of this retrospective study are (1) to determine if a reduction in the conversion rate was obtained when specific measures on the location and path of the LAD artery, as determined from preoperative computed tomography (CT) scans, were available to the surgeon before surgery and (2) to determine if specific patient dimensions can be correlated to the need for conversion to open surgery. Based on the above objectives, we hypothesize that (1) the availability of information on the location of the LAD artery from the center of the sternum on the preoperative CT scans, before robotics-assisted minimally invasive coronary bypass grafting, reduced the chance of conversion to open surgery compared with previous procedures in which the information was not available; and (2) a correlation exists between specific patient dimensions obtained on the CT scan and the success or failure of the minimally invasive robotic procedure.
A retrospective study was performed, including a total of 144 patients who underwent robotics-assisted single vessel LITA-LAD bypass surgery5 during the period of November 2003 to December 2007. Institutional Review Board approval and patient consent for the collection and use of the data were obtained before the procedures being performed. All patients were scanned using CT a few days before the surgical procedure, using a four-slice multidetector CT scanner with a 0.8-second rotation time (GE Medical Systems, Lightspeed, CT/Milwaukee, WI). The CT scans were all performed using the following protocol: enhanced, helical 5-mm collimation with a 3-mm reconstruction interval, 0.75:1 pitch, and a 36-cm field of view. A total of 120 mL of radiographic contrast (Omnipaque 300; Amersham) was injected intravenously using a power injector at a rate of 3 mL per second. Patients were positioned under the direction of the cardiothoracic surgeon to mimic intraoperative positioning. The position of the patient during scanning mimicked that of the actual surgical procedure: 30 degrees left anterior oblique position with a wedge placed below the left scapula and the left arm at the patient's side. If the position of the patient in the CT scanner deviated from the ideal of 30 degrees, corrections to any measurements performed on the images were made before analysis.
Beginning in May 2005, the analysis of the CT scans was modified to include measurement of the LAD artery location. Postprocessing of the images was performed for each patient by a fellowship-trained radiologist. The same radiologist analyzed each patient to avoid interobserver error. The proximal to mid-LAD coronary artery was localized at the left anterior third, fourth, or fifth intercostal space (best-fit location—the standardized point of measurement was the course of the LAD coronary artery between the third and the fifth intercostal space; measurements were taken on the fourth intercostal space, unless it was shown to be intramyocardial; if so, the measurements were taken at the third or the fifth space depending on the best point of access), and the distance from a point on the skin surface located on the vertical axis drawn from the LAD artery to the midline of the sternum was measured (line 1 in Fig. 2). The depth of the LAD artery from the skin surface was also measured (line 2 in Fig. 2). On the same image slice, the thoracic cavity size was measured, and the ratio of the anteroposterior (AP) diameter to the transverse (Trv) diameter of the thoracic cavity was determined. The procedures were rated as unsuccessful if conversion to open surgery was required or as successful if the procedure was completed in a minimally invasive manner.
Statistical analysis of the collected data was performed using the Statistical Package for Social Sciences (SPSS, Chicago, IL) software, version 17.0 for Windows. Analysis of the results included a comparison of success rates before (55 patients) and after May 2005 (89 patients) and a statistical summary of measured data on the 89-patient dataset.
For further analysis, the Shapiro-Wilk test was used for most of the variables to determine whether they were parametric or nonparametric. If the kurtosis of the variable suggested that the outliers had a strong impact on the variable, the extreme outliers were removed from the dataset, and the variable was retested for normality using the Shapiro-Wilk test. If the latter step was unsuccessful, both the extreme and nonextreme outliers were removed and retested for normality using the Shapiro-Wilk test. If the data was determined to be nonparametric, the entire dataset, complete with outliers, was used in the rest of the analysis.
The data were then analyzed to determine the presence of any correlations between the variables. The Pearson correlation coefficient was used to determine the presence of correlation between parametric variables, whereas Spearman's rho was used to evaluate the correlation between parametric and nonparametric data. Dummy variables were used for regression analysis of categorical variables. Numeric data were converted into ordinal data through visual binning with one standard deviation separations. The information was further analyzed to make an in-depth comparison between procedures that failed because of a lack of space and the successful cases.
The comparison analysis shows that there was a reduction in the conversion rate from 31% (in the 55-patient dataset) to 15% (in the 89-patient dataset) when the proposed CT scan analysis was incorporated into practice. Also, there were no cases where a wrong vessel was grafted in the entire series (compared with three cases in a previous series with the ZEUS robot; Computer Motion, Goletta, CA).
The results of the statistical analysis are summarized in Table 1. In this table, the LAD lateral distance (LAD-Lat) refers to the measurement of line 1 in Figure 2, while the LAD depth refers to line 2 in Figure 2. AP diameter is the distance from the anterior border of the vertebral body to the posterior margin of the sternum at the anterior fourth intercostal space. The Trv diameter is the distance from the right to the left pleural margin at the level of the anterior fourth intercostal space. The last column of the table refers to the ratio between the AP diameter and the Trv diameter. This measure is an indication of the overall shape of the thorax, i.e., barrel (value closer to 1) or flat (value closer to 0). Figure 2 shows CT scans of typical barrel-shaped and flat-shaped chests.
To determine if any of the measures had an effect on the success or failure of the procedure, the data were analyzed as a whole and also separated into male and female categories. Correlations significantly differed in each category, as presented in Table 2. There were no significant correlations found between the male data points and the success of the procedure.
These results show that the variable with the strongest significant correlation was LAD-Lat/Trv in the female patients only. Further analysis of these results is presented in Figures 3 and 4. Figure 3 demonstrates a very significant decrease in success rate when the lateral distance is less than 48 mm. Figure 4 demonstrates that an LAD-Lat to Trv ratio of less than 0.20 significantly increases the occurrence of conversion during this procedure in female cases.
There were also some significant correlations found pertaining to the patients with a buried LAD artery and the success of their procedure. Within this set of patients, the lateral distance had the strongest correlation with the success of the operation. Lateral distance had a correlation of 0.347 and a significance of 0.002 at the 0.01 level.
No significant correlations were observed between the male data points and the success of the procedure. Further analysis of the causes of the failed cases revealed that only one patient was converted to open surgery because of limitations in space. Two other patients were converted because of buried vessels, whereas the other patients required conversion because of complications during the procedure. Analysis of the anatomic dimensions of the patient who failed because of insufficient space revealed that most of the dimensions of this patient were quite marginal within the dataset, as shown in Table 3. The measured values can be combined to obtain even more marginal results within the data range, as shown in the bottom rows of Table 3.
A study has been performed to establish evaluation criteria for potential patients for endo-SVST surgery through their preoperative CT scans. The goal was to establish a relationship between the patient dimensions and the success rate of endo-SVST procedures.
The first hypothesis stated that the availability of information on the location of the LAD coronary artery from the center of the sternum on the preoperative CT scans would reduce the chance of conversion to open surgery compared with previous procedures in which this information was not available. The results of the retrospective study show that the conversion rate was reduced from 31% to 15% when the measured data became available to the surgeon before commencing the surgical procedure. Establishing a formal control group was not possible as this was a retrospective study. The reduction in the conversion rate could have also been affected by other factors, such as an increase in clinical experience and a better selection of patients. The greatest benefit of measuring the LAD artery location was that access to the target vessel was achieved through much smaller atraumatic incisions, 3 to 4 cm long. The results also show a significant variability in the location of the LAD artery throughout the dataset (Table 1). This indicates that estimating LAD artery location based on external anatomic landmarks is inaccurate and can lead to improper working angles and restricted reach within the surgical site. This study has shown the importance and effect of properly identifying the location of the LAD coronary artery before a nonrib-spreading incision to optimize working angles, reduce incision size, reduce conversion rates to an open procedure, and reduce errors in grafting to the incorrect vessel. Considering that the analysis was made in a retrospective fashion, it is not possible to isolate the different causes for this reduction.
An additional hypothesis of the study was that it would be possible to find a correlation between the success or failure of the procedure and specific patient dimensions obtained from the preoperative CT scan. Within the female cases, the variable that had the strongest significant correlation with success was the lateral distance from the LAD artery to the sternum (LAD-Lat) divided by the Trv of the patient (0.532, P = 0.006). This is a very promising result and most certainly demonstrates an area of focus for further exploration. However, it should be noted that these cases were not converted to open surgery because of lack of space, but rather because the LAD artery was buried. Therefore, these data suggest that there is a relationship between the patient dimensions and the chance of presenting a buried LAD artery. Furthermore, it cannot be concluded that a ratio between the lateral distance and the transverse distance provides information pertaining to a volume of space that is too small for the surgeon to continue the surgery. It is still interesting to note that of 26 female cases studied, only 3 cases were converted, all 3 because of buried LAD arteries, and 3 of those cases had very similar lateral distances. This indicates that there may be a relationship between the distance from the LAD to the sternum and the chance of conversion because of a decrease in LAD accessibility, which is aggravated in individuals with a small chest cavity such as women.
The same tests were run with the 89 male cases, 10 of which were conversions for several different causes (1 case because of lack of space, 1 case with a buried LAD artery, and 8 cases were converted for different and unforeseeable reasons). Significant correlations between the patient dimensions and success rate were not found in the male cases as their lateral distances were more dispersed. There was one documented male case in this dataset of conversion primarily because of lack of space. Analysis of the data shows that most of the patient dimensions were far from the mean. Combinations of these dimensions could provide a measure of patient eligibility. These results, although not conclusive, indicate that it might be possible to find preoperative CT scan measurements that will lead to the identification of ineligible patients because of space limitations. A larger sample size of failed cases because of limited space is needed to identify which dimensions are important and their effect on the success of the procedure.
The data above cannot prove our second hypothesis; however, several trends in the data have indicated that preoperative measurements can determine the eligibility of a patient for the endo-SVST procedure. Current work is focused on recording relevant data during ongoing procedures to further support these claims. The goal is to find appropriate patient characteristic measurements to provide accurate eligibility criteria for robotic procedures, based on preoperative patient dimensions.
The results presented herein are applicable to endo-SVST procedures, in which the anastomosis is performed through a minithoracotomy. The location of the LAD from the preoperative CT scan to the intraoperative conditions is affected by the collapse of the left lung; however, this small shift in location has not been found to be significant. If the procedure is performed in a totally endoscopic manner, the shift in LAD location would be aggravated by chest insufflation, and additional analysis would need to be performed to determine the effect of this shift.
The need for conversion to an open procedure could be influenced by other significant variables. One that is considered of great importance is that the actual location of the ports and instrument approach angles could be correlated to the measurements collected from the CT scans. A patient with average chest measurements could still undergo conversion if the ports are poorly planned and the instruments cannot reach or have limited maneuverability within the surgical site. Furthermore, limited space within the chest may affect the ability to properly place the ports for optimal access to the surgical site. Further analysis of this correlation is part of ongoing studies.
This study aimed to establish evaluation criteria for potential patients who can undergo endo-SVST surgery based on their preoperative CT scans. This study has shown the importance of identifying the location of the LAD coronary artery before a minithoracotomy to optimize working angles, reduce incision size, reduce conversion rates to an open procedure, and reduce errors in grafting to the incorrect vessel. It was also found that the lateral distance of the LAD coronary artery from the midline divided by the transverse thoracic width of a female patient showed a significant correlation with procedure success. No significant correlations were found for male patients.
1. Kiaii B, McClure RS, Stitt L, et al. Prospective angiographic comparison of direct, endoscopic, and telesurgical approaches to harvesting the internal thoracic artery. Ann Thorac Surg.
2. Vassiliades TA Jr. Endoscopic-assisted atraumatic coronary artery bypass. Asian Cardiovasc Thorac Ann.
3. Vassiliades TA Jr, Douglas JS, Morris DC, et al. Integrated coronary revascularization with drug-eluting stents: immediate and seven-month outcome. J Thorac Cardiovasc Surg.
4. Kiaii B, McClure RS, Stewart P, et al. Simultaneous integrated coronary artery revascularization with long-term angiographic follow-up. J Thorac Cardiovasc Surg.
5. Boyd WD, Kiaii B, Novick RJ, et al. RAVECAB: improving outcome in off-pump minimal access surgery with robotic assistance and video enhancement. Can J Surg.
6. Kiaii B, Rayman R, Swinamer S, et al. Abstract 3051: a novel approach to hybrid coronary artery revascularization using bivalirudin. Circulation.
7. Mitka M. Beat goes on in “off-pump” bypass surgery; surgeon experience may be key to best outcome. JAMA.
8. Ballantyne G. Robotic surgery, telerobotic surgery, telepresence, and telementoring: review of early clinical results. Surg Endosc.
9. Subramanian V, Patel N. Current status of MIDCAB procedure. Curr Opin Cardiol.
10. DelRizzo DF, Boyd WD, Novick RJ. Safety and cost-effectiveness of MIDCABG in high-risk CABG patients. Ann Thorac Surg.
11. Schachner T, Feuchtner GM, Bonatti J, et al. Evaluation of robotic coronary surgery with intraoperative graft angiography and postoperative multislice computed tomography. Ann Thorac Surg.
12. Falk V, Diegeler A, Walther T, et al. Total endoscopic computer enhanced coronary artery bypass grafting. Eur J Cardiothorac Surg.
13. de Cannière D, Wimmer-Greinecker G, Cichon R, et al. Feasibility, safety, and efficacy of totally endoscopic coronary artery bypass grafting: multicenter European experience. J Thorac Cardiovasc Surg.
14. Bonatti J, Schachner T, Bonaros N, et al. Technical challenges in totally endoscopic robotic coronary artery bypass grafting. J Thorac Cardiovasc Surg.
15. Klein P, Meijer R, Eikelaar JHR, et al. Epicardial ultrasound in off-pump coronary artery bypass grafting: potential aid in intraoperative coronary diagnostics. Ann Thorac Surg.
16. Reiter G, Reiter U, Bergmann P, Rienmuller R. MR imaging-based port placement planning for totally endoscopic coronary artery bypass grafting. Interact Cardiovasc Thorac Surg.
This retrospective study from Trejos et al at the University of Western Ontario examines which image measurements from preoperative computed tomography (CT) scans correlated with procedural success during robotically-assisted coronary artery bypass grafting. This group has been a pioneer in this field and has made a number of important contributions. In this study of 144 patients, a strong correlation was found between the success rate and the ratio of the lateral distance to the transverse diameter in female patients. A ratio of less than 0.20 increased the occurrence of conversion during this procedure. There were no significant correlations found for male patients. They also found that information on the location of the left anterior descending coronary artery from the center of the sternum on the preoperative CT scan decreased the conversion rate from 31% to 15%. The main weakness of this study was that there was no control group. Although these results are not conclusive, they do suggest that better preoperative imaging and planning can lead to better surgical results and a reduced conversion rate. Many of the conversions were because of an intramyocardial left anterior descending coronary artery, which may also be identified with newer higher resolution CT scans. The authors are to be congratulated for undertaking this retrospective review of their data and demonstrating the utility of careful preoperative planning. Further work in this area is needed and will hopefully continue to refine and improve the results of minimally invasive endoscopic coronary bypass grafting.
Keywords:© 2010 Lippincott Williams & Wilkins, Inc.
Surgical planning; Robotics-assisted cardiac surgery; Image guidance; Minimally invasive surgery