| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Division of Cardiothoracic Surgery, Oregon Health and Science University, Portland, Oregon
b Department of Surgery, Oregon Health and Science University, Portland, Oregon
Accepted for publication August 21, 2009.
* Address correspondence to Dr Schipper, Division of Cardiothoracic Surgery L353, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd, Portland, OR 97239-3011 (Email: schippep{at}ohsu.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: The Nationwide Inpatient Sample (NIS) was queried for procedure codes for pneumonectomy, lobectomy, limited lung resection, and decortication. We constructed multivariate logistic regression models to calculate odds of hospital mortality or length-of-stay (LOS) greater than 14 days (a marker of morbidity), adjusted for age, sex, patient comorbidities, hospital setting, and surgeon specialty. A surgeon was considered general thoracic if they performed greater than 75% general thoracic operations and less than 10% cardiac operations, Cardiac if greater than 10% cardiac operations, and general surgeon if less than 75% general thoracic and less than 10% cardiac operations. A second set of models additionally adjusted for procedure-specific hospital and surgeon volume.
Results: From 1996 to 2005, the NIS estimates 41,808 pneumonectomies, 321,767 lobectomies, 75,200 limited lung resections, and 149,318 decortications were performed in the United States. For all procedures studied, general thoracic surgeons had significantly decreased odds-of-death and LOS greater than 14 days compared with general surgeons. Cardiac surgeons had significantly decreased LOS greater than 14 days for all operations and decreased odds-of-death for decortications, lobectomy, and limited lung resection compared with general surgeons. When further adjusted for surgeon volume, most differences in odds-of-death were no longer present; however, significantly decreased LOS greater than 14 days largely persisted for both general thoracic and cardiac surgeons.
Conclusions: The majority of general thoracic surgical operations in the United States are performed by surgeons not specializing in thoracic surgery. Both general thoracic surgeons and cardiac surgeons achieve better outcomes than general surgeons. Differences in mortality may be more dependent on surgeon volume than subspecialty. Differences in morbidity are significantly impacted by surgeon specialty and volume.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This study seeks to determine a difference, if any, in the quality of care delivered by general thoracic surgeons, cardiac surgeons, and general surgeons for noncardiac, general thoracic surgical procedures. We use a federally compiled database designed to obtain a nationwide sampling, including all payers, across all non-federal hospitals to determine the associations between surgeon practice patterns and mortality and length of stay after general thoracic surgical operations.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The NIS is the largest all-payer, inpatient care database in the United States. The database is a stratified, cross-sectional sample that includes approximately 20% of all community (non-federal) hospital discharges in the United States. The sampling frame for the 2005 NIS is a collection of hospitals that comprises approximately 90% of all hospital discharges in the United States. To ensure the representative nature of the database, the NIS is stratified by geographic region, urban versus rural location, teaching status, hospital ownership, and hospital bed size. The NIS is managed under the Healthcare Cost and Utilization Project of the Agency for Healthcare Research and Quality and is available to the public for health services research and to policy makers for analysis of national trends in health care utilization, access, charges, quality, and outcomes. It includes discharge information regardless of payer, including persons covered by Medicare, Medicaid, private insurance, and the uninsured. The NIS provides statistical sampling weights allowing for national estimates of patient characteristics and outcomes. For this study we combined data from the 1996 through 2005 NIS databases. Over this time the number of states in the NIS grew from 19 to 37. In 2005, the database contained discharge data on approximately 8 million hospital stays at 1,054 hospitals.
The NIS was searched for hospital discharges of patients age 18 and older that had an International Classification of Diseases, 9th revision, Clinical Modification (ICD-9-CM) procedure code of any of four index general thoracic surgical cases: decortication (34.51), segmental resection of lung-partial lobectomy (32.3 and 32.9), lobectomy (32.4), and pneumonectomy (32.5). The individual surgeons performing these cases were divided into three groups based on percent of their total case volume spent doing general thoracic surgery, cardiac surgery, or any other type of surgery. A surgeon was considered general thoracic if greater than 75% of total cases performed were from ICD-9-CM codes considered general thoracic and no more than 10% of total cases performed were from ICD-9-CM codes considered cardiac. A surgeon was considered cardiac if greater than 10% of their cases were considered cardiac. A surgeon was considered general surgeon if fewer than 75% of total cases were general thoracic and fewer than 10% of total cases were cardiac. Note that board certification and subspecialty training were not considered in these definitions as this information was not available in the NIS. The definition of surgeon type is based on the practice pattern for each surgeon. Also, because surgeon identifiers were not necessarily consistent from year to year, all surgeons were given year specific identifiers and were classified based on that year's case mix. The ICD-9-CM codes used for the definitions of cardiac surgeon and general thoracic surgeon are available by contacting the authors.
We constructed a multivariate logistic regression model to calculate a patient's odds of hospital mortality and a second model to calculate odds of length of stay (LOS) greater than 14 days. Mortality was in-patient only. Length of stay greater than 14 days is used as a surrogate marker for morbidity in the manner of Wright and colleagues [5], developed from the Society of Thoracic Surgeons (STS) General Thoracic Database [5]. Wright and colleagues noted that patients with a LOS greater than 14 days had 3.2 complications per patient compared with 0.4 complications per patient for those with a shorter LOS. In addition they noted that 14 days is a true statistical outlier for LOS after lobectomy, eliminating any disagreement about the severity of a complication sustained.
Our first set of multivariate models adjusted for age, sex, patient comorbidities [6], hospital setting, and surgeon specialty. Specific patient comorbidities and hospital settings adjusted for are listed in Table 1. Age was adjusted as a categoric variable (18 to 50, 51 to 70, 71 to 80, and 81 or older). A second set of multivariate models was then constructed to calculate odds of death and odds of LOS greater than 14 days, additionally adjusting for procedure specific hospital volume and procedure specific surgeon volume. In these models, both hospital and surgeon volume were included as continuous, logarithmically transformed variables. The logarithmic transformation helps normalize the distribution of volumes and corresponds to an interpretation that the change in outcome is associated with a multiplicative increase (say, doubling) in volume. All models account for the sampling structure of the NIS including sampling weights to allow national estimates, clustering of cases within hospitals, and the structure of the stratified sample.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The majority of all four operations were performed by surgeons whose practice consisted of less than 75% general thoracic and who did not do cardiac surgery. Surgeons whose practice consisted of 75% or greater of procedures considered general thoracic and less than 10% procedures considered cardiac, our definition of general thoracic surgeon, performed the fewest of the index procedures (Table 2).
|
|
|
The second multivariate analysis was constructed to evaluate predictors for risk of LOS greater than 14 days. The odds of having a LOS greater than 14 days for all four index procedures in the analysis not considering procedure specific surgeon and hospital volume were consistently lowest for general thoracic surgeons compared with cardiac surgeons and general surgeons (Table 4). The difference between general thoracic surgeons and general surgeons was statistically significant for all four procedures. While general thoracic surgeons consistently had lower odds of death than cardiac surgeons in this model, the model was not designed to detect a difference between these two groups. The odds of a having a LOS greater than 14 days were statistically different and lower for all four procedures performed by cardiac surgeons compared with general surgeons. When procedure specific hospital volume and procedure specific surgeon volume were added to the morbidity model, most of the significant differences in odds of morbidity persisted. General thoracic surgeons had statistically lower odds of LOS greater than14 days for pneumonectomy, limited lung resection, and decortication, but similar odds of LOS greater than 14 days for lobectomy compared with general surgeons. Cardiac surgeons had statistically lower odds of LOS greater than 14 days for all four index procedures compared with general surgeons. See Table 4 for details of the multivariate analysis calculating odds of morbidity with and without procedure specific volumes.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
However, the risk of death and the risk of morbidity for four thoracic operations were decreased if those operations were performed by a surgeon whose practice was devoted primarily to cardiothoracic surgery (both cardiac and general thoracic) as opposed to general surgery. The reduction in morbidity and mortality was even greater if those procedures were performed by a surgeon whose practice was devoted solely to general thoracic surgery. Although these differences were partially mitigated by adjustment for procedure specific volume, much of the benefit persisted especially with regard to morbidity.
In some aspects, our findings are consistent with previous studies examining this issue, in others they are not. Using an administrative database collected by the State of South Carolina, Silvestri and colleagues [3] showed that American Board of Thoracic Surgery (ABTS) certified thoracic surgeons had a significantly lower mortality rate for lobectomy over surgeons board certified in general surgery (3.0% vs 5.3%). This difference in mortality between specialties was also seen in patients with extreme comorbidities (25.4% vs 43.6%) and age greater than 65, (3.5% vs 7.4%). Using Medicare data from 1998 to 1999, Goodney and colleagues [1] showed ABTS certified surgeons had a significantly lower mortality rate for lung resection (lobectomy or pneumonectomy) compared with general surgeons (5.8% noncardiac thoracic surgeons, 5.6% cardiothoracic surgeons, 7.6% general surgeons). Furthermore, they found that in surgeons performing more than 20 lung resections annually, mortality rates were lowest for ABTS certified noncardiac thoracic surgeons (5.1% noncardiac thoracic, 5.2% cardiothoracic, and 6.1% general surgeons). This difference held for surgeons at high volume hospitals [1]. Goodney and colleagues [1] report an adjusted operative mortality for lobectomy by noncardiac thoracic surgeons, cardiothoracic surgeons, or general surgeons of 4.9%, 4.8%, and 6.5%, and for pneumonectomy of 13.1%, 13.1%, and 17.2%, respectively. The mortalities we found for these groups were 2.3%, 3.0%, and 4.1% for lobectomy and 6.4%, 10.1%, and 11.5% for pneumonectomy. Our results may have been in part influenced by the difference in patient populations between the NIS and Medicare databases. The NIS is a nationwide sampling of a cohort that contains adults of all ages and insurance statuses, including the uninsured. When this nationwide sample is considered, only 50% of hospital discharges have Medicare as a payer and can be expected to be included in Medicare data. The NIS includes an additional 37% of discharges with private insurance and 13% that are self pay or unknown not represented in Medicare data. Possibly because the NIS includes younger patients the mortality rates for lobectomy and pneumonectomy are lower in the NIS than in the Medicare databases. We also noted that many significant differences in mortality but not in morbidity between surgeon types were no longer present when hospital and surgeon procedure specific volume was considered. It may be that the low rate of mortality with general thoracic procedures, especially in younger patients, makes this a poor measure of discerning differences in outcomes. Morbidity may be a more useful measure.
The four case types evaluated were chosen as a representative sample of the surgical treatment of both benign and malignant diseases of the chest and to include procedures that historically have been performed by general surgeons, general thoracic surgeons, or cardiothoracic surgeons. Esophagectomy and esophageal surgeries were considered general thoracic procedures for purposes of the definition of general thoracic, cardiac, and general surgeon; however, we chose not to include esophagectomy as an index case for this study. Dimick and colleagues [2] found that specialty training in thoracic surgery had an independent association with lower mortality after esophageal resection, but was less important than hospital and surgeon volume. Rogers and colleagues [7] utilized the NIS database to further evaluate the effect of surgeon, volume, and hospital on mortality after esophagectomy. They found that patient factors had a greater influence on outcome than individual surgeons and hospitals.
Our definition of a general thoracic surgeon, cardiac surgeon, and general surgeon is based on the practice pattern of the surgeon in a given year, rather than on board certification. It was based on the percentage of all cases that a surgeon performed that were general thoracic and cardiac cases. The NIS database codes each surgeon with a unique, but anonymous, identifier. Therefore, unlike the Medicare database, in the NIS it is not possible to learn the names of individual surgeons and link them to ABTS data. Previous studies examining this issue have used ABTS certification as part of the definition of a general surgeon versus a cardiothoracic surgeon [1–3]. While this strategy allows differentiation between general surgeons and cardiothoracic surgeons as a group, it does not permit one to discern general thoracic surgeons from cardiac surgeons. The ABTS does not differentiate between general thoracic and cardiac surgeons when certifying surgeons and prior to July 2007 there was only one set of case requirements for ABTS certification. Beginning in July 2007, the ABTS recognized two tracks to certification, general thoracic and cardiac, with different case requirements. However, the examinations and the certifications do not indicate which track a candidate completed. Some surgeons complete general thoracic fellowships after ABTS certification, but there is no standardization of, certification for, or centralized tracking of this additional training. For the period of time of our study we felt the practice pattern definition best. In simplest terms, you are what you do. The ABTS certification is an important measure of a safe thoracic surgeon. After formal training the experience a surgeon accrues may be a better measure of their current competency. The ABTS does require the submission of case logs as part of recertification.
The percentages chosen to define cardiac, general thoracic, and general surgeons were designed with the idea that a general thoracic surgeon is someone doing exclusively general thoracic surgery. Our definition required a general thoracic surgeon to have a practice of 75% or greater cases considered general thoracic surgery and no more than 10% cardiac surgery. The remainder of the practice could consist of any case type. A higher qualifying percentage may inadvertently exclude surgeons truly doing only general thoracic surgery due to miscoding of procedures or due to our inability to develop a complete list of general thoracic surgical ICD-9-CM codes. A cardiac surgeon is defined as a surgeon doing 10% or more cardiac cases. Previous authors have used the performance of one coronary artery bypass graft a year as defining a cardiac surgeon from a general thoracic surgeon [1]. We felt this method of identifying a cardiac surgeon could potentially overcount due to miscoding. We do agree with the spirit of the definition in that if you perform at least a few cardiac cases you are subspecialty trained in cardiac surgery. A further study could be conducted to see if there is a threshold of general thoracic case percentages over or under which morbidity or mortality become statistically significant. It is also possible that the discrepancies between our findings and previous studies are because of this difference in definition.
Our analysis has several limitations. First, the NIS is a sample and not a complete database of all hospital discharges. As a result, although it is designed to be representative of national practice, there is the possibility for error. However, the NIS is the largest collection of real data that can currently be used for addressing the present question. Second, the NIS is an administrative database. Administrative data were designed for claims data collection and billing, not health care research, and can be limited by erroneous coding of procedures and comorbidities [8, 9]. However, given the large sample size, random miscoding would be unlikely to significantly influence our findings. Any potential impact would likely bias the results toward the null. Finally, administrative data may fail to collect procedure specific and clinically relevant clinical variables such as functional status and pulmonary function. What these variables are and their impact on thoracic surgery is being evaluated in clinical databases such as the Society of Thoracic Surgery General Thoracic Database. However, until participation in these clinical databases is widespread or nonvoluntary (as is the NIS), they will not be able to address the same questions as this manuscript.
Notwithstanding these limitations, our study was conducted using a large, national dataset with adequate power to generate current, stable mortality and morbidity rates. Such data provide an unbiased, national picture of the practice of surgery. By including information from both high and low performing and high and low volume surgeons and hospitals, they can be used to evaluate the practice of surgeons and hospitals that are less likely to participate in large, voluntary, clinical databases.
We used hospital LOS greater than 14 days as a surrogate for morbidity. This surrogate has been verified for lobectomy [5]. To allow easier comparisons between index surgeries, we also applied this surrogate to pneumonectomy, limited lung resection, and decortication. These applications have not been verified but we considered this reasonable as the basic assumptions in using LOS greater than 14 days for lobectomy are true for pneumonectomy and limited lung resection. The median LOS in the NIS during this time span was 7 days (Q1, Q3: 5, 11) for pneumonectomy, 7 days (Q1, Q3: 5, 11) for lobectomy, and 7 days (Q1, Q3: 5, 11) for limited resection of the lung. A LOS greater than 14 days for these three procedures would be an outlier and a reasonable surrogate for morbidity. For decortication, LOS greater than 14 days may not be a strong measure of severe morbidity or capture outliers as median LOS in the NIS was 14 days (Q1, Q3: 9, 21).
A final consideration is that although our study reveals a reduction in mortality and morbidity of the index procedures if the procedure is performed by a general thoracic surgeon or cardiac surgeon, we have not defined why that reduction occurs. Multiple reasons could be hypothesized which make the procedure safer performed by general thoracic surgeons and cardiac surgeons including specialty training, familiarity with the anatomy and disease processes, or more practice using the techniques involved. It could also be system-based practices that are assembled around or tend to be assembled by general thoracic or cardiothoracic surgeons including specialty wards, nursing, protocols, or assisting or consulting medicine colleagues. Further studies are needed to better define these characteristics and determine what course of action is needed to make these procedures safer for the larger population of patients in the United States who are undergoing general thoracic procedures by surgeons not specializing in thoracic surgery.
The majority of general thoracic surgical operations in the United States are performed by surgeons not specializing in thoracic surgery. Both general thoracic surgeons and cardiac surgeons achieve better outcomes than general surgeons. The lowest rates of mortality and morbidity (LOS > 14 days) were achieved by surgeons whose practice was more than 75% dedicated to general thoracic surgery.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The ACGME [Accreditation Council for Graduate Medical Education] 2006 to 2007 General Surgery National Resident Report reveals that the average lung resections done by general surgery residents during five years is six. Now, I don't know about you, but I want my surgeon to have had more experience. Do you have any data on the ages or duration of practice of general surgeons in the NIS [Nationwide Inpatient Sample] and whether that correlates with outcome? Do you hypothesize that younger general surgeons are learning from senior partners? Unfortunately, nobody can tell us how general surgeons differ from CT [cardiothoracic] surgeons in such critical matters as staging, uses of thoracoscopic techniques, referral for adjuvant treatment, et cetera. These are areas where real differences may exist between your three groups, even general thoracic surgeons versus cardiac surgeons. Based on your findings, would you recommend that thoracic surgery organizations and health policy agencies state that all patients requiring major lung surgery should be referred to CT surgeons?
Thank you. I really enjoyed your paper.
DR SCHIPPER: Thank you. Your first question I can't answer, and the reason for that is the Nationwide Inpatient Sample; while you can bring the data down to a specific surgeon, that surgeon is coded by an anonymous identifier. So other than knowing that a discharge belonged to a surgeon, you know nothing more about that surgeon.
I thought long and hard about your second question, and ultimately what it comes down to is I am able to point out that there is a difference in the type of care and the quality of care that is delivered, but I can't tell you why that difference exists. I had a calculation, and I won't share it because I think it is unfair, but I had a calculation of what are the additional lives lost if you had a general thoracic surgeon or a cardiac surgeon doing this procedure versus a general surgeon. Well, the volume that is being done by general surgeons is so large that I can't tell you that that volume would be able to be assumed by general thoracic surgeons and cardiac surgeons and maintain the same level of care. I personally think it could, but this data can neither support or refute that statement.
DR BRYAN F. MEYERS (St Louis, MO): Nice job on your presentation, but a lot of the results depend on the surgeon classification system that you created since the surgeons weren't classified in the database as being cardiac, thoracic, or general surgeons. You picked something that seems reasonable, but I wondered how you came upon those particular case cutoffs or percentage cutoffs? Did you explore any other classification strategies, and if so, what were your impressions based on those explorations?
DR SCHIPPER: The classification is based on two things, and I think the next paper is going to be varying those classifications and seeing if that makes a difference. Is there a cutoff, 75%, 50%, 60%, where these things become significant or nonsignificant? The classification in this paper was based on, one, if you are doing any cardiac surgery in the list that we defined as cardiac surgeon, then you are most likely a cardiac surgeon. Previous papers have used the definition that if you do one CABG [coronary artery bypass grafting] a year, then you are a cardiac surgeon. We thought that that definition, because there were so many data points in this database, was maybe a little too broad, and we would include people in there just because of miscoding that should not be. So we chose 10%; if 10% or more of your cases are cardiac, then you are a cardiac surgeon. The 75% value was chosen, because if you start to get higher, 80%, 90%, again, you may start to lose some of the people in the database that really are doing the majority of their cases general thoracic but they are doing something else that we inappropriately left off our list. They are doing maybe some vascular surgery or a lot of bronchoscopy, and we didn't want to set the general thoracic standard too high that we ended up with even a smaller group of general thoracic. The 75% value may actually be a little high. In previous papers by Birkmeyer [1], they listed the percent of lobectomies being done by general thoracic surgeons as about 30%, whereas we found it was 10%. So our criteria in comparing to those two papers may actually be stricter.
DR ROBERT J. CERFOLIO (Birmingham, AL): Is the percent of procedures being done by general thoracic surgeons increasing over the time frame of this study, is it staying the same, or is it decreasing?
DR SCHIPPER: I don't know, but I am going to look at it. I thought of that last night.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  D. E. Wood Editorial Comment: What is most important in improving outcomes after pulmonary lobectomy: the surgeon or the approach? Eur J Cardiothorac Surg, April 1, 2013; 43(4): 817 - 819. [Full Text] [PDF]
|
|
|
|
 |
|
 R. R. Gopaldas, C. M. Bhamidipati, T. K. Dao, and J. G. Markley Impact of Surgeon Demographics and Technique on Outcomes After Esophageal Resections: A Nationwide Study Ann. Thorac. Surg., March 1, 2013; 95(3): 1064 - 1069. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. T. Cooke and D. H. Wisner Who Performs Complex Noncardiac Thoracic Surgery in United States Academic Medical Centers? Ann. Thorac. Surg., October 1, 2012; 94(4): 1060 - 1064. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kernstine A Lobectomy by Any Other Name J. Clin. Oncol., August 10, 2012; 30(23): 2803 - 2804. [Full Text] [PDF]
|
|
|
|
 |
|
 N. M. Rueth, H. M. Parsons, E. B. Habermann, S. S. Groth, B. A. Virnig, T. M. Tuttle, R. S. Andrade, M. A. Maddaus, and J. D'Cunha Surgical treatment of lung cancer: Predicting postoperative morbidity in the elderly population J. Thorac. Cardiovasc. Surg., June 1, 2012; 143(6): 1314 - 1323. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. A. Ferraris, S. P. Saha, D. L. Davenport, and J. B. Zwischenberger Thoracic Surgery in the Real World: Does Surgical Specialty Affect Outcomes in Patients Having General Thoracic Operations? Ann. Thorac. Surg., April 1, 2012; 93(4): 1041 - 1048. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. S. Park, F. C. Detterbeck, D. J. Boffa, and A. W. Kim Impact of Hospital Volume of Thoracoscopic Lobectomy on Primary Lung Cancer Outcomes Ann. Thorac. Surg., February 1, 2012; 93(2): 372 - 379. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. Ellis, B. S. Diggs, J. T. Vetto, and P. H. Schipper Intraoperative Oncologic Staging and Outcomes for Lung Cancer Resection Vary by Surgeon Specialty Ann. Thorac. Surg., December 1, 2011; 92(6): 1958 - 1964. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Lim, M. Beckles, C. Warburton, and D. Baldwin Cardiopulmonary exercise testing for the selection of patients undergoing surgery for lung cancer: friend or foe? Thorax, October 1, 2010; 65(10): 847 - 849. [Full Text] [PDF]
|
|
|
|
|
|
P. H. Schipper and B. S. Diggs Reply. Ann. Thorac. Surg., September 1, 2010; 90(3): 1063 - 1063. [Full Text] [PDF]
|
|
|
|
|
|
F. A. Baciewicz Jr and D. Edelman Are There Enough General Thoracic Surgeons? Ann. Thorac. Surg., September 1, 2010; 90(3): 1062 - 1063. [Full Text] [PDF]
|
|
|
|
|
|
D. E. Wood and F. Farjah Surgeon Specialty Is Associated With Better Outcomes: The Facts Speak for Themselves Ann. Thorac. Surg., November 1, 2009; 88(5): 1393 - 1395. [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 |
