| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Surgery, William Beaumont Hospital, Royal Oak, Michigan
b Research Institute, William Beaumont Hospital, Royal Oak, Michigan
Accepted for publication April 2, 2009.
* Address correspondence to Dr Chmielewski, 3577 W 13 Mile Rd, #301, Royal Oak, MI 48073-6769 (Email: gchmielewski{at}beaumont.edu).
Presented at the Fifty-fifth Annual Meeting of the Southern Thoracic Surgical Association, Austin, TX, Nov 5–8, 2008.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: Data were retrospectively collected on patients who underwent pulmonary lobectomy at a single health system between August 1, 2003, and April 1, 2008. Age, sex, pulmonary function tests, preoperative and postoperative stages, pathologic diagnosis, anatomic resection, extent of lymph node sampling, surgical technique and duration, complications, blood loss, transfusion requirement, chest tube duration, length of hospital stay, 30-day readmission, and mortality rate were examined. The percentage of patients who underwent thoracoscopic lobectomy and their outcomes were then compared among three chronologic cohorts.
Results: Three hundred sixty-four patients underwent pulmonary lobectomy (239 open; 99 thoracoscopic; 26 thoracoscopic converted to open). Baseline characteristics, staging, pathologic diagnosis, and anatomic resections were similar in the early, middle, and late cohorts. The percentage of thoracoscopic lobectomies increased from 16% to 49%, whereas open lobectomy decreased from 81% to 42% (p < 0.0001). The complication rate remained constant with the exception of air leaks lasting more than 7 days (9% versus 10% versus 2%; p = 0.02). Hospital length of stay (6 versus 5 versus 4 days; p < 0.0001) and chest tube duration (4 versus 3 versus 3 days; p < 0.0001) decreased and operative duration increased as more thoracoscopic lobectomies were performed. Blood loss, transfusion requirement, 30-day readmission, and 1-year survival were not significantly different among chronologic cohorts.
Conclusions: Established thoracic surgeons can safely incorporate thoracoscopic lobectomy with no increase in morbidity or mortality.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
As evidence supporting minimally invasive surgery mounts and more patients request VATS, thoracic surgeons not formally trained in VATS lobectomy will be challenged to incorporate minimally invasive pulmonary resections into their practice. There are few reports in the literature addressing whether the transition from open to VATS lobectomy can be safely performed. We hypothesized thoracic surgeons without formal minimally invasive training could incorporate this operation into their practices without sacrificing safety or oncologic efficacy. This study retrospectively reviews the outcomes during 2 surgeons' transition from open to VATS lobectomy and details their learning process.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Bronchoscopy was performed and double-lumen single-lung ventilation was used on all patients. Positioning was lateral decubitus with thoracoscopic resections done by the surgeon in the anterior position. A 5-mm, 30-degree Olympus (Orangeburg, NY) camera was used. One to three port sites were constructed as needed with a 4- to 6-cm utility incision in the third or fourth intercostal space. No rib spreading occurred. Wedge resections were performed to confirm pathologic diagnosis if no preoperative tissue diagnosis was available. Next, the pulmonary ligament was released, and the pulmonary vein, artery, and bronchus were individually isolated and ligated with linear staplers. Fissures were completed last, and all specimens were retrieved with an endoscopic bag (catalog no. G15656; Cook Lapsac Surgical Tissue Pouch, Bloomington, IN). Lymph node harvesting was performed throughout the procedure for staging and to facilitate dissection of vascular structures. Early in the surgeons' experience, minimally invasive attempts at resection were reserved for patients with favorable anatomy (nonobese habitus, complete fissures, no adhesions, peripherally based tumors, <3 cm in size, minimal lymphadenopathy). As experience was gained, all patients deemed to have early-stage non–small cell lung cancer (NSCLC) were approached for minimally invasive resection. Patients received a single 28F chest tube at the conclusion of the case. Either epidural or patient-controlled analgesia was used in the postoperative period. Patients were admitted to the surgical intensive care unit overnight and transferred to the surgical floor the following morning.
Time limits were set for each step of the operation, and if not met, the case was converted to open. Cases were classified as a conversion if a vascular structure or bronchus was divided before the decision to proceed with thoracotomy. For patients converted secondary to bleeding, data on intraoperative vasopressor use and blood product transfusion was recorded. In addition, pathologic stage and diagnosis, postoperative complications, blood transfusions, chest tube duration, hospital length of stay, 30-day readmission rate, and mortality were collected for each patient. All mortality data were queried from the Social Security Death Index and is current through May 20, 2008. When data were not available, no assumptions were made and the number of patients analyzed is reported. Patients who underwent pulmonary lobectomy during the surgeons' early (defined as first one third of lobectomies), middle (second one third), and late (third one third) experience were then compared for differences in the collected variables. In addition, patients who underwent open lobectomy were compared with those who underwent VATS lobectomy for all variables using an intention-to-treat analysis (conversions included with VATS).
Categorical variables were examined using a 
2 test when appropriate (expected frequency > 5); otherwise Fisher's exact test was used. All continuous variables were examined using the Kruskal-Wallis test, as none of these variables were normally distributed. Median and 90th percentile are reported for these variables. Kaplan-Meier survival curves were generated to compare the early, middle, and late cohorts at 1 year. In addition, Kaplan-Meier survival curves were used to compare patients who underwent lobectomy with the VATS and open techniques. The log-rank test was used, and a probability value of less than 0.05 was considered statistically significant. SAS version 9.1.3 (SAS Institute, Cary, NC) was used for all analyses.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
When patients who underwent VATS or conversion lobectomy were independently examined, sex distribution, spirometry values, pathologic diagnoses, preoperative and postoperative staging, distribution of anatomic resection, and number of lymph nodes harvested were not significantly different among cohorts. The median patient age decreased slightly from the early to late cohorts (Table 2). There was no difference in N1 or N2 nodes collected between the open and VATS techniques in any of the time periods, with the exception of the early period when a median of one more N2 node was harvested with open lobectomy (Table 3).
|
|
|
|
|
|
|
There were no intraoperative or 30-day mortalities in this series. According to Kaplan-Meier analysis, patients in the early, middle, and late cohorts had equivalent survival at 1 year (Fig 2). In addition, when all patients were examined, those who underwent VATS or conversion lobectomy had equivalent survival to those who underwent open lobectomy (Fig 3). Similarly, when survival was examined in NSCLC patients alone, stage I NSCLC patients alone, and stage II NSCLC patients alone, no difference was observed between the minimally invasive and open techniques.
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Recent reports have concluded that VATS lobectomy can be safely taught to trainees [15] and suggested that training be coordinated at the national level to concentrate experience and facilitate learning [16]. However, until this occurs, the question remains whether thoracic surgeons, experienced in performing open lobectomy and basic thoracoscopy, can safely incorporate VATS lobectomy into their practice? Our data suggest that this can be accomplished without compromising patient outcomes. In a patient population that did not change, the percentage of VATS lobectomies performed tripled while open lobectomies decreased proportionally. During this transition, the number of patients with air leaks lasting greater than 7 days, chest tube duration, and hospital length of stay all decreased.
The observed reduction in chest tube duration and hospital length of stay was partially attributable to the increasing proportion of VATS lobectomies performed as a function of time. However, chest tube duration and hospital length of stay each decreased within the open group as a function of time. This trend is likely attributable to changes in the intraoperative and postoperative management of thoracotomy patients as a result of our VATS experience. Adaptations such as the increased use of stapling devices and reduction in electrocautery to complete fissures may have resulted in the observed decrease in air leaks after open lobectomy. The surgeons also tended to remove chest tubes with higher outputs (up to 350 mL/day) in the late cohort than in the early cohort (up to 250 mL/day), contributing to the overall decreased chest tube duration and hospital length of stay. In addition, early in their experience, the surgeons began using intraoperative 0.5% ropivacaine intercostal nerve blocks and intravenous patient-controlled analgesia for selected VATS patients. The transition from intravenous patient-controlled analgesia to oral pain medication was well tolerated and more expeditious than from epidural analgesia to oral medication. As a result, 1 of the surgeons began selectively using intraoperative intercostal nerve blocks and intravenous patient-controlled analgesia for thoracotomy patients, potentially contributing to the observed reduction in hospital length of stay among the open lobectomy patients.
Although our study was not designed to address the superiority of VATS versus open lobectomy, our median hospital length of stay was shorter in the late VATS group than the late open group. Even though changes in the management of the open patients reduced the hospital length of stay as a function of time, it never dropped below a median of 5 days. Conversely, patients in the late VATS lobectomy group had a median hospital stay of 4 days, suggesting that once VATS lobectomy is learned, the technique is at least equivalent in this respect. The chest tube duration and rate of air leaks greater than 7 days were also equivalent between the late open and late VATS groups. These data support our hypothesis that thoracic surgeons without formal minimally invasive training can learn thoracoscopic lobectomy without compromising patient outcomes.
The largest increase in VATS lobectomies occurred between the middle and late cohort, indicating that proficiency with the technique occurred after approximately 50 cases. This is comparable to the number of cases required for proficiency with advanced laparoscopic skills, such as laparoscopic prostatectomy [17]. In our opinion, the main challenges are the change in visualization, restricted tissue manipulation, and decrease in tactile feedback when performing VATS dissections. The surgeon's view changes from the traditional posterior-lateral thoracotomy position to an anterior-caudal thoracoscopic view. To make this transition easier, one may consider using an anterior thoracotomy and dissecting structures from the anterior position during open lobectomies. Before each case, a video review of the procedure is recommended for both surgeon and assistant. This reinforces the view of the lobar anatomy and proper dissection planes. In addition, to gain comfort with the loss of tactile sensation, a conscious effort can be made to perform open cases with sharp and electrocautery dissection only.
Adopting VATS pulmonary resection may not be a fit for all thoracic surgeons. Both surgeons in this series had basic thoracoscopic and laparoscopic skills and an extensive experience with pulmonary resection through the posterior lateral thoracotomy. In addition, they led their institution's multidisciplinary thoracic oncology teams and recognized the need for accurate staging and coordination of multispecialty care. A graded, sequential approach was used to introduce the VATS lobectomy into their practice. The primary goal was to duplicate the open lobectomy, and established guidelines for VATS resections were strictly adhered to [18]. Early in their experience, only nonobese patients who had small (<3 cm) peripheral lesions, a negative mediastinoscopy, complete fissures, minimal adenopathy, and no neoadjuvant therapy were attempted thoracoscopically. As experience was gained, nearly all patients were approached with a minimally invasive intent.
An adequate pace of resection should be maintained. Recommended sequences of dissections have been published for the five basic pulmonary resections. When starting out, reasonable time limits should be set to reach these anatomic goals (generally 20 to 30 minutes per structure), and if not met, the surgeon should convert to an open approach. On opening, inspection of the area of difficulty often provides insight to be used in future cases. In our early cohort, the median VATS operative duration was approximately 1 hour longer than the median open case. Counterintuitively, there was no reduction in VATS operative duration as experience was gained with minimally invasive techniques. This is likely a reflection of the increasingly complex lesions addressed with VATS as comfort was gained with minimally invasive techniques. With experience, more patients with incomplete fissures, prior thoracic or cardiac surgery, obese body habitus, dense adhesions, and benign lymphadenopathy were attempted thoracoscopically. The threshold at which to perform a thoracotomy was determined intraoperatively and changed with experience. However, patients with T3 or T4 lesions and those with tumors greater than 5 cm in size, as well as a few early patients with a body mass index in excess of 35, were not attempted thoracoscopically. Most recently, we began performing VATS lobectomies in patients after neoadjuvant therapy and VATS segmentectomies.
Our overall conversion rate from VATS to thoracotomy was 21%. Half of our conversions were for anatomic factors; examples included a left upper lobe truncus anterior that was tight on the bronchus, a buried right upper ascending artery, and bronchi that we could not obtain proper staple angles on. The consistent rate of conversion from the early to late cohorts reflects the surgeons' learning curve as more difficult lobectomies were attempted, as well as a consistent threshold at which conversion occurred. As described previously, conversion for bleeding was rarely dramatic. Only 7% (9 of 125 patients) of patients were converted because of hemorrhage in our series. Most bleeding can be controlled with application of a sponge stick for tamponade while the case is converted to an open approach. A single patient in the early cohort required intraoperative transfusion of blood products as a result of hemorrhage. This patient ultimately did well, with no measurable morbidity.
There are a multitude of resources available to gain the knowledge and the skill set required to perform minimally invasive pulmonary resections. One innovation was a multiauthored paper, presented much like an industry tech manual, helpful in troubleshooting VATS resections [19]. Video recordings of specific anatomic resections can be useful for learning the technique, orienting oneself to the anterior-caudal view of the anatomy, and monitoring progress relative to other surgeons. Surgeons have worked with industry to distribute video compact discs demonstrating various anatomic resections [20]. In addition, videoconferencing has allowed physicians to view live minimally invasive pulmonary resections during conferences, and site visits to high-volume centers can be very beneficial in refining techniques.
The low number of N2 lymph nodes collected in this series likely reflects our practice of selective lymph node sampling. Every patient in this series underwent preoperative CT scan and PET scan, or integrated PET-CT scan later in the series. In addition, all patients underwent mediastinoscopy and biopsy, unless all mediastinal lymph nodes were less than 1 cm on CT scan and no uptake was demonstrated on PET scan. Once stations 2, 4, and 7 were sampled by mediastinoscopy, we did not routinely return to these stations to biopsy lymph nodes during lobectomy. This practice may be criticized given the potential for false-negative cervical mediastinoscopy. However, the majority of positive N2 lymph nodes found at lobectomy are not accessible by mediastinoscopy [21]. Furthermore, our survival data do not suggest that our patients were understaged. We had an approximate 90% 2-year survival and did not find that a group of stage IA patients dropped off our survival curves early, as would be expected if a group of stage IIIA patients were misstaged as IA. Finally, if we explored a station and did not visualize any lymph nodes, we did not routinely harvest a fat pad specimen from that station. In reviewing our operative notes, 49 patients (17, 13.6% VATS conversion; 32, 13.4% open) had at least one station explored without nodes being present.
The clinical significance of the median of one additional node harvested during open lobectomy in the early group is uncertain, as our data do not demonstrate a survival difference between the VATS and open patients. After approximately 20 VATS lobectomies, the number of nodes harvested with VATS was equivalent to that of the open lobectomy. We believe VATS lymph node sampling can be performed as effectively as that done through thoracotomy, particularly as experience is gained. Our data suggest that the nodal dissection must not be neglected while learning the VATS lobectomy. When beginning to perform the procedure, particular attention should be given to adequate nodal sampling.
Early in our experience, patients were counseled that the goal of the operation was to achieve the same oncologic outcome whether it was through thoracotomy or VATS. It was explained that the surgeons had extensive experience with the open procedure and basic thoracoscopic skills; however, this was a new procedure, and patient safety would ultimately guide the operation. Therefore, although the operation would be started with minimally invasive intent, the patient may receive a thoracotomy. After their questions were answered, all patients who were offered attempt at VATS lobectomy wished to proceed.
As evidence supporting minimally invasive pulmonary resection grows and patients increasingly request VATS, thoracic surgeons will be faced with the prospect of learning VATS lobectomy. By continually challenging the limits of our thoracoscopic abilities, with an utmost respect for patient safety and a willingness to convert to thoracotomy when appropriate, we have successfully incorporated VATS lobectomy into our practice. Our data suggest that with time, VATS lobectomy can be integrated into a thoracic surgeon's practice without compromising patient outcomes.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
You reported a conversion rate of 21%, and I would like to congratulate your group on presenting that. Oftentimes in the literature it is much lower, and I think early in one's experience this is a true representation of what the conversion rate should be. It is not a failure. It is actually just a component of good judgment in the setting of a learning curve. Can you please elaborate on the anatomic factors that contributed to conversion? Can you also comment on the bleeding factors that made you convert? Were they semielective, semiurgent, or was there a fair bit of kicking and screaming and gnashing of teeth. Also how can we prevent these episodes and how can we treat them when they do occur?
Point two. In your results, your air leak, your length of stay, and your chest tube duration decreased over the three time periods. However, it appears that the most significant change over time occurred in the open group rather than the VATS group. How did the institution of VATS change your intraoperative and postoperative strategy management of these thoracotomy patients? Clearly, there had to be a change in your practice. Along these lines, in the third or late group, your chest tube duration was the same as well as the length of stay between the VATS and open. Does this argue that the two techniques are similar? Can you therefore present some information that would convince Dr Cerfolio that VATS is superior to a well-managed muscle-sparing, rib-sparing, nerve-sparing and now chest tube–sparing thoracotomy?
Third, the median number of N2 lymph nodes in both groups was 1, which seems somewhat small. Can you comment on your group's approach to lymph node dissection versus sampling? Furthermore, there was a difference in that more N2 lymph nodes were dissected with the thoracotomy group compared to the VATS. Can you suggest why this is the case in your paper, which is in contrast to the Japanese studies? I would also comment that with a median number of 1, there had to be a fair number of patients that had no N2 lymph nodes dissected, which I would suggest may not be adequate.
Finally, the VATS took about 1 hour more than the open technique, but I noticed over your three time periods that you didn't decrease your operative time with the VATS technique, despite the fact that there was a similar percentage of 1A patients in all three time periods. Can you elaborate why with increasing experience you were not able to decrease the operative time? Furthermore, can you make some suggestions to surgeons contemplating doing VATS whether or not to dive directly into VATS or to do some sort of hybrid procedure?
Thank you.
DR SEDER: Thank you, Dr Crabtree, for your insightful comments and questions. You asked about our 21% VATS to open conversion rate. This number being higher than the generally quoted rate of 1% to 11% reflects our willingness to open and dedication to patient safety, as well as a distinct effort to keep an adequate pace throughout the operation. We set goals for the completion of the dissection and ligation of structures; a rough estimate might be a half hour per structure. If we did not meet these goals, we converted to open. One half of our conversions were for anatomic factors; that is what we have in our manuscript. Examples might include a left upper lobe truncus anterior that is tight on the bronchus or a buried right upper ascending artery or perhaps a bronchus that we just could not get the correct staple angle on. Nine out of our 26 conversions were done so for bleeding. All of these except one, when the PA (pulmonary artery) was torn, were for what we considered nuisance bleeding. As experience has been gained, we have become better at controlling this type of bleeding with pressure and perhaps a piece of fibular, moving to another area, and then coming back to the bleeding. For torrential bleeding, which is always a concern in a thoracoscopic case, we always have a sponge stick on our stand for controlling bleeding while we perform a thoracotomy.
You asked why our rate of prolonged air leak, chest tube duration, and length of stay all decreased in the open groups as experience was gained with VATS surgery. We absolutely changed our practice pattern with regard to open cases when we saw how well our VATS patients were doing. Intraoperatively, we converted from the use of blunt dissection and electrocautery dissection of the fissures to stapling. When we saw that the only thing keeping them in the hospital on postoperative day 1 was a chest tube that had 350 mL of output, we were more aggressive in removing chest tubes. So, applying our VATS experience to our open experience absolutely changed our practice pattern.
You inquired into why our median number of N2 lymph nodes harvested was 1 in the open and VATS groups. We generally perform lymph node sampling, not a full lymphadenectomy. Nearly every one of these patients underwent a preoperative CT (computed tomography) scan, PET (positron emission tomography) scan, and mediastinoscopy. Once stations 2, 4, and potentially 7 had been sampled by mediastinoscopy, we generally did not return to these areas in order to sample lymph nodes at lobectomy. So, we might take an azygos node or perhaps a paraesophageal node, and that may be partially what is reflected in the low number of N2 lymph nodes that you see here.
Of note, we do not feel that these patients were understaged either. Even though our median follow-up was only 2 years, we did not see a lot of stage 1A patients dropping off on the Kaplan-Meier curve as you would expect if those truly were stage 3A patients.
Next you asked why our VATS operative times had not dropped and continued to hover right around 200 minutes in the early, middle, and late cohorts in the face of any equal number or potentially even an increasing number of 1A patients in the late cohort. While these numbers are all true, factors that were not be measured and examined included pleural adhesive disease, incomplete fissures, prior thoracic operations, degree of lymphadenopathy, and obese body habitus. We would have been more likely to open in the presence of such factors early in our experience than late in our experience. Therefore, persistence with the thoracoscope likely contributed to the equivalent operative durations seen later in our experience.
Finally, you asked if we used a hybrid operation while we were learning. None of the VATS patients in this series had an incision larger than a 6-cm utility incision. However, we did take a few steps early in our experience to help make the transition from open to VATS lobectomy easier. This included transitioning from the traditional posterolateral approach to an anterior muscle-sparing approach. This gave us a feel for the anterior view, which you use during your VATS lobectomy. In addition, on occasion, if we weren't comfortable continuing on with VATS, we would create a thoracotomy, leave the scope in, and continue our VATS dissection. Again, patient safety was our first concern. In addition, we noted that the loss of tactile sensation is one of the most difficult parts of learning the VATS lobectomy; in an effort to overcome this, when we were doing our open lobectomies, we tried to perform them with electrocautery and sharp dissection, somewhat eliminating our reliance on tactile sensation.
DR ROBERT J. CERFOLIO (Birmingham, AL): So your false-negative rate for mediastinoscopy must be 0, is that correct?
If you don't go back and get the nodes after you do a med, then you must assume your med is perfect every time and thus you have a false-negative rate of 0 for your med. I wish mine was that low. My point is I suggest that just because you have done a med, that you should still go back and get those nodes and you will be surprised how many people have an endoscopic ultrasound (EUS) and an endobronchial ultrasound (EBUS) and a mediastinoscopy that we are told "prove the patient is N2 negative" and yet they still have N2 disease. They only way to prove it is to remove all the nodes whether they were biopsied or not. Then you have to ask yourself, does that make a difference in their outcome, and if they get adjuvant chemo does it help this select group of patients. I think it may. So in my practice, and although there is no definitive data for that, I think it is better to go back and do a complete thoracic lymphadenectomy even if you have done a mediastinoscopy, or an EBUS or an EUS. And is it scarred? Yes, but that is the job, I think, to remove not only the pulmonary malignancy but all the N2 mediastinal lymph nodes as well.
DR SEDER: I appreciate your comment, and I have to assume that our false-negative rate is not 0. However, as stated earlier, our Kaplan-Meier survival curves do not suggest that our patients were understaged. Meaning, the patients that we staged as 1A did not act like stage 3A patients over time.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. C. Tong, M. R. Gustafson, S. S. Balderson, T. A. D'Amico, and S. L. Meyerson Validation of a thoracoscopic lobectomy simulator Eur J Cardiothorac Surg, February 6, 2012; (2012) ezs012v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Amer, A.-Z. Khan, H. Vohra, and R. Saad Is it safe to include octogenarians at the start of a video-assisted thoracic surgery lobectomy programme? Eur J Cardiothorac Surg, February 1, 2012; 41(2): 346 - 352. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. C. Giulianotti, N. C. Buchs, G. Caravaglios, and F. M. Bianco Robot-assisted lung resection: outcomes and technical details Interact CardioVasc Thorac Surg, October 1, 2010; 11(4): 388 - 392. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Paul, N. K. Altorki, S. Sheng, P. C. Lee, D. H. Harpole, M. W. Onaitis, B. M. Stiles, J. L. Port, and T. A. D'Amico Thoracoscopic lobectomy is associated with lower morbidity than open lobectomy: a propensity-matched analysis from the STS database. J. Thorac. Cardiovasc. Surg., February 1, 2010; 139(2): 366 - 378. [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 |
