| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2006;82:164-171
© 2006 The Society of Thoracic Surgeons
Division of Cardiothoracic Surgery, Doernbecher Children's Hospital, Oregon Health and Science University, Portland, Oregon
Accepted for publication March 3, 2006.
* Address correspondence to Dr Welke, Division of Cardiothoracic Surgery L353, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd, Portland, OR 97239-3098 (Email: welkek{at}ohsu.edu).
Presented at the Fifty-second Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 10–12, 2005.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
METHODS: We requested data for all pediatric cardiac surgical procedures performed between 2001 and 2004 at 29 Congenital Heart Surgeon's Society (CHSS) member institutions (using CHSS as a surrogate for recognized high quality). Procedures were categorized by Risk Adjustment for Congenital Heart Surgery, version 1 (RACHS-1) category. In-hospital mortality rates for each category were calculated and compared with those in the 2002 manuscript of Jenkins and colleagues.
RESULTS: We received data for 16,805 procedures from 11 institutions. In all, 12,672 operations (76%) could be placed into RACHS-1 categories. Overall in-hospital mortality for categorized operations was 2.9% and was most related to case mix. There was a significant decrease in the percentage of category 1 operations, and there were significant increases in category 2, 4, and 6 operations. There were significant decreases in category 2, 3, 4, and 6 mortality rates (Jenkins 2002 [CHSS]): (1) 0.4% [0.7%], (2) 3.8% [0.9%], (3) 8.5% [2.7%], (4) 19.4% [7.7%], (5) not applicable, and (6) 47.7% [17.2%]. There was no significant association between hospital surgical volume and mortality.
CONCLUSIONS: This outcomes "footprint" suggests that we could hold ourselves accountable to higher benchmarks than those reflected by some published standards. Mortality rates declined, despite an increase in case mix complexity. The lack of association between hospital surgical volume and mortality suggests that other factors determine outcomes at high-quality institutions. In addition to continually validating our expectations for treatment, future research needs to identify these factors by understanding the system of care and identifying process measures that influence outcomes.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
In 2002, Jenkins and colleagues [1] put forth a risk-stratification system for congenital cardiac surgery. The Risk Adjustment for Congenital Heart Surgery (RACHS-1) method was created to allow a refined understanding of differences in mortality among patients undergoing congenital heart surgery [2]. As part of this manuscript, multi-institutional mortality rates for congenital heart surgery, derived from 1994 to 1996 data, were reported. Although these mortality rates were calculated primarily for validation of the risk adjustment methodology, they are commonly used by centers as benchmarks for comparing their own results, both internally for quality improvement and externally for patient education and advertising. However, these mortality rates may not represent standards to be attained in current practice, since much has changed in congenital cardiac surgery over the past 10 years leading to improved outcomes. Additionally, they were obtained from a wide range of hospitals performing congenital heart surgery, not only centers with a specific commitment to treating children with congenital heart disease.
The purpose of our investigation is to evaluate whether these published and widely used standards for pediatric cardiac surgery accurately reflect current expectations. In addition, we seek to determine the strength of RACHS-1 category and hospital volume as predictors of mortality. Our hypotheses are that (1) mortality rates at high-quality pediatric cardiac programs are lower than published national results despite (2) a change in case mix with a shift away from low complexity operations. We also hypothesize that, unlike RACHS-1 category, hospital volume is a poor discriminator of mortality.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Procedures were categorized using the simple form of the RACHS-1 method. This approach groups procedures with similar expected short-term mortality rates into 6 risk categories. Category 1 has the lowest risk of death and category 6 the highest (Table 1). Operations were categorized either by the surgeons at the respective centers, or by the lead author. A detailed description of the risk adjustment methodology is in the original manuscript [1].
|
Data used for calculation of the original standard were obtained from two sources, the Pediatric Cardiac Care Consortium (PCCC) database for calendar year 1996 and hospital discharge data purchased from three states (Illinois 1994, Massachusetts 1995, and California 1995) [1]. The PCCC collected and analyzed data for all cardiac procedures performed at member institutions on an annual basis for quality improvement purposes. The PCCC data for cardiac surgical procedures performed on patients under 18 years of age at 32 institutions were used for the analyses. In all, 4,602 cardiac procedures were included in this dataset, 4,370 (95%) of which could be assigned to a RACHS-1 category. The hospital discharge data contained information on 4,493 patients, 3,646 (81%) of which could be assigned to a RACHS-1 category.
For each institution in our study, yearly and overall case mix and unadjusted in-hospital mortality rates were calculated. To maintain consistency with the report of Jenkins and colleagues, in-hospital mortality was defined as death occurring in the same hospitalization as the surgical procedure regardless of when after surgery the death occurred. The numerator was the number of deaths and the denominator was the number of operations. The overall case mix and mortality rate for the cohort was calculated as well. Mortality rates for each of the 6 RACHS-1 categories were also calculated for each institution and the entire cohort. Comparisons of case mix and mortality rates between centers and data sets were made using 95% confidence intervals.
The discrimination of RACHS-1 category as a predictor of mortality was assessed by the c statistic [3, 4] as determined from univariate logistic regression analysis. The c statistic is numerically equivalent to the area under a receiver operator characteristic curve and represents the proportion of times that a given patient who died had a higher predicted risk of death (on the basis of RACHS-1 category in this case) than a given patient who lived. The c statistic ranges from 0.5 to 1.0 and has been widely reported for multivariable models predicting mortality for patients undergoing cardiac surgery. A test with perfect discrimination has a c statistic of 1.0. A test with no discrimination has a c statistic of 0.5, in other words, a coin flip. To maximize potential discrimination, the RACHS-1 categories were modeled using indicator variables, in lieu of a single ordinal variable.
Next, the relationship between hospital volume and mortality was examined. Hospital volume was defined as the average number of RACHS-1 categorized procedures performed per year over the 4 years of the study. First, volume was evaluated as a continuous variable. Second, hospital volume was categorized into terciles by selecting whole-number division points that most closely sorted the sample into three relatively equal size hospital cohorts: low (fewer than 200 a year), medium (200 to 300 a year), and high (more than 300 a year). Several approaches were used to define the relationships between hospital volume and mortality. First, unadjusted mortality rates across volume groups were compared using the 
2 statistic for linear trend. Second, the discrimination of volume alone as a predictor of mortality was assessed by the c statistic as determined from univariate logistic regression analyses for the two approaches for analyzing hospital volume described above. Multivariable analysis examined the increase in model discrimination (namely, c statistic value) resulting from the addition of hospital volume to a model containing RACHS-1 category as an independent predictor. The model was adjusted to account for patients clustered within hospitals. For all analyses, the procedure was the unit of analysis, and in-hospital mortality was chosen as the dependent variable. All analyses were conducted using STATA statistical software (Version 7.0; STATA Corporation, College Station, Texas).
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Compared with the 1996 PCCC dataset, there was a shift away from RACHS-1 category 1 operations. There were slight increases in category 2, 4, and 6 operations (Table 2). Compared with the 1994 to 1995 hospital discharge dataset, the trends were similar except for no change in the frequency of category 2 operations and a slight increase in category 3 operations. The number of operations in category 5 was low; therefore, category 5 was omitted from further analyses.
|
Individual institutional overall mortality ranged from 1.0% to 6.0%. Mortality rates for category 1 (median, 0.0; range, 0.0 to 3.1) and category 2 (median, 0.8; range, 0.0 to 1.9) were low, with five centers having no deaths in category 1 and two centers having no deaths in category 2. There was slightly more variability in category 3 mortality rates (median, 3.0; range, 1.0 to 3.9), with one center outperforming the group mean. Category 4 (median, 5.6; range, 0.0 to 18.2) mortality rates differed more, but owing to wider confidence intervals (secondary to lower numbers of operations) only one center performed better than the group mean. The greatest variation was seen in category 6 mortality (median, 16.7; range, 1.2 to 48.8). In this category, one center outperformed and one center underperformed the group mean (Fig 1). When centers were ranked based on mortality rates for each RACHS-1 category, no center was consistently the best or worst performer (Fig 2).
|
|
Hospital volume had poor discrimination for predicting mortality with a c statistic of 0.55. When divided into terciles, the discrimination of volume remained poor (c statistic of 0.55). Hospital volume did not contribute significantly to the predictive value of a multivariate model containing RACHS-1 category and adjusted for clustering within center. In addition, the ability of hospital volume by RACHS-1 category to predict mortality for each category, such as the ability of category 4 volume to predict category 4 mortality, was also poor.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
We chose to use the RACHS-1 method of risk adjustment to maintain consistency with the widely quoted standard from the original report. The RACHS-1 method was not developed to determine operative risk for individual patients, but to compare the mortality for groups of patients undergoing congenital cardiac surgery, such as variability among centers. Certain procedures that occur less commonly are not included in the RACHS-1 method. However, low frequency procedures are less important when comparing overall center performance. In addition, the methodology does not allow for specific conditions that may be risk factors for specific operations. However, such potential risk factors are likely to be evenly distributed in large populations. Despite these limitations, the RACHS-1 method is a widely used and rigorously validated risk adjustment methodology for congenital heart surgery [1, 6–9].
There was a higher percentage of unclassified operations in our cohort than in the datasets used by Jenkins and colleagues. That may be because centers in our cohort submitted their full operative experience, including not only major cardiac operations but also thoracic operations and more minor procedures performed by congenital cardiothoracic surgeons. Operations in the Jenkins study were identified by PCCC and International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) operative and procedure codes indicating surgical repair of a congenital heart defect. In addition, the Jenkins study excluded certain groups of patients, including those 30 days of age or younger who had repair of an isolated patent ductus arteriosis and those who underwent cardiac transplantation. The more recently developed Aristotle basic complexity score for congenital heart surgery categorizes a higher percentage of operations, in part due to improvements in nomenclature [9]. Planned modifications to RACHS-1 include mapping of the operations in the risk categories to the International Pediatric and Congenital Cardiac Code nomenclature to address this issue. For our study, however, placement of these operations in the unclassified category was appropriate to maintain consistency with the original report.
The lower mortality rate seen in our cohort may be due to several factors. First, mortality rates for congenital heart surgery have decreased in the past 10 years as surgical techniques and systems of care have improved. Second, our cohort differed from the PCCC and hospital discharge cohorts in that, instead of being a broad representation of hospitals likely to be of varying volume and quality, it was a group of hospitals that were presumed to be committed to excellence in congenital heart disease based on surgeon membership in the CHSS.
In contrast to the decrease in mortality in all other categories, there was no significant change in category 1 mortality. Concurrently, there was a decrease in the percentage of category 1 operations. The change in case mix may be due to innovations in interventional cardiology. Many patients undergoing the three most prevalent category 1 operations (atrial septal defect closure, closure of patent ductus arteriosus in patients more than 30 days of age, and coarctation repair in patients more than 30 days of age) are now treated with interventional techniques rather than surgery. The lack of decrease in mortality may be due to the same phenomenon. Lower risk patients with less complex lesions are treated interventionally whereas higher risk patients with more complex lesions are treated with surgery.
While the lower mortality rates achieved by our cohort suggest that previously published standards should not serve as current benchmarks, the performance of our cohort may not be an appropriate benchmark either. A benchmark is a point of reference against which other things are compared or measured. It is a goal to be attained. In congenital cardiac surgery, mortality benchmarks should be set at levels consistent with or better than the best performers in our field. They should be criterion based rather than norm based so that they are theoretically achievable by all centers. A benchmark differs in purpose from the mean plus or minus two standard deviations, which is appropriate for describing current levels of care and identifying outliers. Even though the performance of our cohort was superior to previous reports, there was still variability in mortality rates. A better benchmark would be obtained by examining all centers and determining the mortality rates of the top performing group.
The variability in mortality rates between centers is an opportunity for education and improvement. While a portion of the variability seen may be attributed to chance, the remainder is likely the result of center-specific differences in care. Across RACHS-1 categories, no one center was consistently the best performer, indicating that all centers could learn from the practice of others. Sharing of information regarding practice patterns and systems of care through site visits and regional and national quality improvement initiatives would result in better outcomes for all centers.
Previous studies have shown an association between hospital volume and mortality [10–12]. We did not see such a relationship in our cohort. This may be attributed to the fact that previous investigations have included hospitals with a broader range of surgical volumes and a broader range of recognized quality. The lack of association between hospital surgical volume and mortality suggests that other factors determine outcomes at high-quality institutions. These factors are apt to be process measures and characteristics of the systems of care not currently captured in either administrative or most clinical databases (adherence to known best practices, preoperative preparation, experience and ability of all members of the patient care team, teamwork, postoperative care).
Our analysis has several limitations. First, the surgeons who performed the operations submitted the data. That introduces the possibility of false reporting and bias; however, the range of mortality rates across centers supports their integrity. Second, the only patient-related independent risk factor in our risk models was RACHS-1 category. However, the discrimination of the RACHS-1 method in our study was consistent with previous reports [1, 7–9]. The inclusion of additional clinical risk factors (age, presence of major noncardiac structural anomaly, prematurity) would have added slightly to the predictive value of the model, but would likely not have changed or perhaps diminished further the lack of effect of volume. Third, mortality rates were calculated as deaths per procedures rather than deaths per patients. Although this choice may have impacted a small number of patients who died after multiple procedures, it was made to maintain consistency with the original manuscript. Fourth, our cohort was a select group of centers, which may not be representative of the general population of congenital cardiac surgery programs and, in addition, may not represent a true benchmark. Unfortunately, at present no perfect database or risk adjustment methodology exists. Miscoding and a limited ability to adjust for severity of illness restrict the utility of administrative databases. Currently available clinical databases offer superior risk adjustment and more accurate coding, but contain select samples rather than broad, national coverage. A more appropriate standard could be determined from a national clinical database with full participation of all centers. In the future, the Society of Thoracic Surgeons Congenital Heart Surgery Database may be useful in this role [13, 14]. Further refinements to the RACHS methodology and the Aristotle complexity based score [15] will result in more accurate tools for risk adjustment and evaluation of quality of care. Notwithstanding these limitations, our study was conducted using a large, multicenter dataset with adequate power to generate current, stable mortality rates.
This outcomes "footprint" suggests that we could hold ourselves accountable to higher standards than those reflected by some published benchmarks. Mortality rates declined, despite an increase in case mix complexity. The lack of association between hospital surgical volume and mortality suggests that other factors determine outcomes at high-quality institutions. Importantly, the results from this study are achievable by all centers if the system characteristics and processes of care that contribute to high-quality care can be recognized and put into practice. In addition to continually validating our expectations for treatment, future research needs to identify these factors by understanding systems of care and identifying process measures that influence outcomes. Future risk-stratification systems should focus not only on mortality, but also morbidity, neurologic outcomes, and functional status. Dissemination and implementation of the knowledge gained will benefit patients everywhere.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
This first slide is a photograph of the incoming President of the Southern Thoracic Surgical Association, Ross Ungerleider, shown with a tiny 2-inch fish that he caught. Ross has volunteered here to demonstrate this fishing equivalent of a complexity level 1 operation. This second slide is the current Southern Thoracic President, Irv Kron, fly-fishing with me in Tampa Bay and catching the 12-inch fishing equivalent of a complexity level 2 operation. This third slide is a photograph taken on a fishing trip organized by a the Southern Thoracic Past President, Gus Mavroudis, and it shows Marshall Jacobs and his wife on my boat collaborating to catch a 18-inch Crevalle Jack, the fishing equivalent of a complexity level 3 operation. This fourth slide was created when I traveled out to Oregon to collaborate with Ross to catch this 40-inch Pacific Salmon, the fishing equivalent of a complexity level 4 operation. Finally, our incoming President Ross sent me this picture of his 5-foot salmon, the fishing equivalent of a complexity level 5 or 6 operation.
Now, seriously, your manuscript presents some very important work, and the STS Congenital Database Taskforce has also embraced this concept of complexity adjustment for multi-institutional outcomes analysis. We have used Aristotle methodology for complexity adjustment, and this next slide shows how, similar to the RACHS-1 system, the four levels of the Aristotle system demonstrate excellent correlation between complexity and mortality. This final slide shows the same concept using the entire Aristotle basic complexity score based on an analysis of more than 27,000 STS cases and again demonstrates an excellent correlation between complexity and mortality.
From your manuscript and from these data, I can draw two extremely important conclusions, which will then lead to two questions I have for you. First, it is clear that any methodology of complexity adjustment will need to be adjusted over time. Techniques and technologies improve. The complexity scoring systems must evolve to keep pace with these improvements in surgical technique and biomedical technology. So, my first question is, "how often do you think a complexity adjustment tool should be modified to keep pace with these changes in medicine?" Second, and most important, I think that the most critical lesson from your research is that all cardiac surgeons and all cardiac programs should participate in and submit data to the STS database. And thus, my second question is, "do you have any ideas or strategies to increase participation in the STS database?"
DR WELKE: Thank you, Dr Jacobs, for your kind remarks. Regarding your first question, the influence of specific risk factors and the outcomes of our treatments constantly change. A risk model should be able to respond to these changes, yet it also has to be practical to maintain and use. Updating risk models is time-intensive. In addition, these models are often incorporated into software problems like the STS database, and updating software is also time-intensive and expensive. So we compromise and choose a cycle of updating that moderates cost while maintaining accuracy and relevance. The STS and the EACTS have chosen a 4-year cycle for their updating.
Regarding your second question, the Northern New England Cardiovascular Disease Study Group was the first to monitor cardiac surgery outcomes on a regional level with an eye toward quality improvement. This effort was not started wholly as an altruistic effort but rather in part as a response to governmental threat to end payment at hospitals where mortality rates were too high.
We are now at a point where payers are examining and trying to decipher our pediatric cardiac surgery outcomes. We as surgeons have the best understanding of our specialty and should lead this effort rather than end up being judged by erroneous measures. For example, the State of Oregon this year began reporting pediatric cardiac surgery volumes at the two hospitals in our state that perform such surgery, and the Web site includes no discussion of case mix. The best way for all of us to get involved and influence our future is to, one, maintain a personal database so we know our own results, and two, participate in the STS national cardiac database. This clinical database can then be used to report outcomes that are more relevant and more accurate than those obtained from administrative data. Governmental agencies and payers are starting to ask the questions, and we need to be proactive in providing the right answers because they will impact our and our patients' futures.
DR CHRISTO I. TCHERVENKOV (Montreal, PQ): It is a real privilege to be attending for the first time the meeting of the Southern Thoracic Surgical Association. Although I spend at least a month in Florida every year, I guess I don't qualify for membership because I live and work in Montreal. Perhaps the rules for membership could be changed to include those that spend more than a month per year in the South.
Karl, congratulations for your very important work! One question I would like to ask regards the issue of volume and outcome. You have said that institutional volume does not relate to outcome. This means to me that probably there are complex processes that are involved in the quality of care and outcomes. So my question is, since you have found no correlation between institutional volume and outcome, have you looked at the following relationships: volume per surgeon and outcome, volume within each risk level and outcome, volume within each risk level per surgeon and outcome?
I think there is a lot more work that has to go into the assessment of the relationship of volume and outcome. This is an important area to clarify, because government agencies and payers and others are making the erroneous assumption that if you simply have a larger institutional volume you are going to have better outcome. We know that this is not necessarily the case.
DR WELKE: Thank you, Dr Tchervenkov. We looked at the volume-mortality relationship in a number of ways including by RACHS-1 category and found the ability of hospital volume by category to predict mortality to also be poor. We did not look at surgeon specific volume, as the data were submitted by institution without surgeon identifiers. The problem with our data set for looking too much into volume is that it contained only 11 centers. And, unlike the report by Kathy Jenkins and also that by Ed Hannon using the New York data, we had a much tighter range of volumes in our group. Our smallest hospital had an annual volume of about 100 cases per year. That, in addition to the overall high quality of the institutions in our data set, may be a reason why we did not see much of a volume–mortality relationship. So it is not to say that volume may not be associated with mortality in all cases, but it certainly was not here.
The real problem with volume is that it is a surrogate for quality, not a measure of quality in and of itself. What we really need to do is move beyond volume and look at the process measures and the other structural measures that account for the apparent contribution of volume to outcome.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Mosca Balloon atrial septostomy let's take a closer look. J. Am. Coll. Cardiol., May 12, 2009; 53(19): 1812 - 1813. [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. Jacobs, R. J. Cerfolio, and R. M. Sade The Ethics of Transparency: Publication of Cardiothoracic Surgical Outcomes in the Lay Press. Ann. Thorac. Surg., March 1, 2009; 87(3): 679 - 686. [Full Text] [PDF]
|
|
|
|
|
|
K. F. Welke, B. S. Diggs, T. Karamlou, and R. M. Ungerleider Comparison of Pediatric Cardiac Surgical Mortality Rates From National Administrative Data to Contemporary Clinical Standards Ann. Thorac. Surg., January 1, 2009; 87(1): 216 - 223. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Redlin, A. Koster, M. Huebler, W. Boettcher, N. Nagdyman, R. Hetzer, H. Kuppe, and W. M. Kuebler Regional differences in tissue oxygenation during cardiopulmonary bypass for correction of congenital heart disease in neonates and small infants: relevance of near-infrared spectroscopy. J. Thorac. Cardiovasc. Surg., October 1, 2008; 136(4): 962 - 967. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R.-K. R. Chang, M. Gurvitz, and S. Rodriguez Missed Diagnosis of Critical Congenital Heart Disease Arch Pediatr Adolesc Med, October 1, 2008; 162(10): 969 - 974. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. F. Welke, B. S. Diggs, T. Karamlou, and R. M. Ungerleider The Relationship Between Hospital Surgical Case Volumes and Mortality Rates in Pediatric Cardiac Surgery: A National Sample, 1988-2005 Ann. Thorac. Surg., September 1, 2008; 86(3): 889 - 896. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Shillingford, M. M. Glanzman, R. F. Ittenbach, R. R. Clancy, J. W. Gaynor, and G. Wernovsky Inattention, Hyperactivity, and School Performance in a Population of School-Age Children With Complex Congenital Heart Disease Pediatrics, April 1, 2008; 121(4): e759 - e767. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M J E Bradbury, G Taylor, P Short, and M D Williams A comparative study of anticoagulant control in patients on long-term warfarin using home and hospital monitoring of the international normalised ratio Arch. Dis. Child., April 1, 2008; 93(4): 303 - 306. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. V. Johnston Congenital Heart Disease and Brain Injury N. Engl. J. Med., November 8, 2007; 357(19): 1971 - 1973. [Full Text] [PDF]
|
|
|
|
 |
|
 L. Mildh, V. Pettila, H. Sairanen, and P. Rautiainen Predictive value of paediatric risk of mortality score and risk adjustment for congenital heart surgery score after paediatric open-heart surgery Interactive CardioVascular and Thoracic Surgery, October 1, 2007; 6(5): 628 - 631. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. C. Odegard, J. A. DiNardo, B. D. Kussman, A. Shukla, J. Harrington, A. Casta, F. X. McGowan Jr, P. R. Hickey, E. A. Bacha, R. R. Thiagarajan, et al. The Frequency of Anesthesia-Related Cardiac Arrests in Patients with Congenital Heart Disease Undergoing Cardiac Surgery Anesth. Analg., August 1, 2007; 105(2): 335 - 343. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. Guillemaud, H. El-Hakim, S. Richards, and N. Chauhan Airway Pathologic Abnormalities in Symptomatic Children With Congenital Cardiac and Vascular Disease Arch Otolaryngol Head Neck Surg, July 1, 2007; 133(7): 672 - 676. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. H. Jones The Year in Cardiovascular Surgery J. Am. Coll. Cardiol., May 8, 2007; 49(18): 1887 - 1898. [Full Text] [PDF]
|
|
|
|
 |
|
 P. S. McQuillen, A. J. Barkovich, S. E.G. Hamrick, M. Perez, P. Ward, D. V. Glidden, A. Azakie, T. Karl, and S. P. Miller Temporal and Anatomic Risk Profile of Brain Injury With Neonatal Repair of Congenital Heart Defects Stroke, February 1, 2007; 38(2): 736 - 741. [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 |
