Publications by authors named "Liqi Feng"

14 Publications

  • Page 1 of 1

Society of Thoracic Surgeons 2021 Adult Cardiac Surgery Risk Models for Multiple Valve Operations.

Ann Thorac Surg 2021 Apr 9. Epub 2021 Apr 9.

Duke University, Durham, North Carolina.

Background: The STS Quality Measurement Task Force has developed risk models and composite performance measures for isolated coronary artery bypass grafting surgery (CABG), isolated aortic valve replacement (AVR), isolated mitral valve replacement or repair (MVRR), AVR+CABG, and MVRR+CABG. To further enhance its portfolio of risk-adjusted performance metrics, STS has developed new risk models for multiple valve operations +/- CABG procedures.

Methods: Using July 2011 to June 2019 STS Adult Cardiac Surgery Database (ACSD) data, risk models for AVR+MVRR (n=31,968) and AVR+MVRR+CABG (n=12,650) were developed with the following endpoints: operative mortality, major morbidity (any one or more of the following: cardiac reoperation; deep sternal wound infection/mediastinitis; stroke; prolonged ventilation; and renal failure), and combined mortality and/or major morbidity. Data were divided into development (July 2011 - June 2017, n=35,109) and validation (July 2017 - June 2019, n=9,509) samples. Predictors were selected by assessing model performance and clinical face validity of full and progressively more parsimonious models. Performance of the resulting models was evaluated by assessing discrimination and calibration.

Results: C-statistics for the overall population of multiple valve +/- CABG procedures were 0.7086, 0.6734, and 0.6840 for mortality, morbidity, and combined mortality and/or morbidity in the development sample, and 0.6953, 0.6561, and 0.6634 for the same outcomes, respectively, in the validation sample.

Conclusions: New STS-ACSD risk models have been developed for multiple valve +/- CABG operations, and these models will be used in subsequent STS performance metrics.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.athoracsur.2021.03.089DOI Listing
April 2021

Postoperative Hematocrit and Adverse Outcomes in Pediatric Cardiac Surgery Patients: A Cross-Sectional Study From the Society of Thoracic Surgeons and Congenital Cardiac Anesthesia Society Database Collaboration.

Anesth Analg 2021 Mar 15. Epub 2021 Mar 15.

Department of Anesthesiology and Critical Care Medicine, Bloomberg Children's Hospital, Johns Hopkins University School of Medicine, Baltimore, Maryland.

Background: We sought to examine potential associations between pediatric postcardiac surgical hematocrit values and postoperative complications or mortality.

Methods: A retrospective, cross-sectional study from the Society of Thoracic Surgeons Congenital Heart Surgery Database (STS-CHSD) and Congenital Cardiac Anesthesia Society Database Module (2014-2019) was completed. Multivariable logistic regression models, adjusting for covariates in the STS-CHSD mortality risk model, were used to assess the relationship between postoperative hematocrit and the primary outcomes of operative mortality or any major complication. Hematocrit was assessed as a continuous variable using linear splines to account for nonlinear relationships with outcomes. Operations after which the oxygen saturation is typically observed to be <92% were classified as cyanotic and ≥92% as acyanotic.

Results: In total, 27,462 index operations were included, with 4909 (17.9%) being cyanotic and 22,553 (82.1%) acyanotic. For cyanotic patients, each 5% incremental increase in hematocrit over 42% was associated with a 1.31-fold (95% confidence interval [CI], 1.10-1.55; P = .003) increase in the odds of operative mortality and a 1.22-fold (95% CI, 1.10-1.36; P < .001) increase in the odds of a major complication. For acyanotic patients, each 5% incremental increase in hematocrit >38% was associated with a 1.45-fold (95% CI, 1.28-1.65; P < .001) increase in the odds of operative mortality and a 1.21-fold (95% CI, 1.14-1.29; P < .001) increase in the odds of a major complication.

Conclusions: High hematocrit on arrival to the intensive care unit (ICU) is associated with increased operative mortality and major complications in pediatric patients following cardiac surgery.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1213/ANE.0000000000005416DOI Listing
March 2021

Action Recognition Using a Spatial-Temporal Network for Wild Felines.

Animals (Basel) 2021 Feb 12;11(2). Epub 2021 Feb 12.

College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China.

Behavior analysis of wild felines has significance for the protection of a grassland ecological environment. Compared with human action recognition, fewer researchers have focused on feline behavior analysis. This paper proposes a novel two-stream architecture that incorporates spatial and temporal networks for wild feline action recognition. The spatial portion outlines the object region extracted by Mask region-based convolutional neural network (R-CNN) and builds a Tiny Visual Geometry Group (VGG) network for static action recognition. Compared with VGG16, the Tiny VGG network can reduce the number of network parameters and avoid overfitting. The temporal part presents a novel skeleton-based action recognition model based on the bending angle fluctuation amplitude of the knee joints in a video clip. Due to its temporal features, the model can effectively distinguish between different upright actions, such as standing, ambling, and galloping, particularly when the felines are occluded by objects such as plants, fallen trees, and so on. The experimental results showed that the proposed two-stream network model can effectively outline the wild feline targets in captured images and can significantly improve the performance of wild feline action recognition due to its spatial and temporal features.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/ani11020485DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7917733PMC
February 2021

Costs Associated with Lobectomy for Lung Cancer: a novel analysis merging STS and Medicare data.

Ann Thorac Surg 2020 Nov 11. Epub 2020 Nov 11.

Background: Costs related to care of patients who undergo lobectomy for lung cancer may vary depending on patient, disease, and treating facility characteristics. We aimed to identify underlying case mix factors that contribute to variability of 90-day costs of lobectomy for early stage lung cancer.

Methods: The Society of Thoracic Surgeons General Thoracic Surgery Database was queried for lobectomy for clinical stage I lung cancer (2008-2013). Demographics, clinical outcomes, and 90-day episode-of-care costs across all care settings were analyzed for patients successfully linked to Medicare data. Hospital costs were estimated from charges using cost-to-charge ratios. Comprehensive regression models were created to identify impact of preoperative patient factors and hospital characteristics on costs, and to delineate additive costs due to perioperative outcomes and complications.

Results: The mean 90-day cost for lobectomy was $45,080 ±$38,239. Variables associated with significant additive costs were age ≥75 years, ASA III or IV, FEV1 <80% predicted, BMI <18.5 or >35, current or past smoker, cerebrovascular disease, chronic kidney disease, impaired functional status, open thoracotomy, prolonged operative time, government hospitals, metropolitan setting, and geographic location. Patients with ≥1 postoperative complication resulted in an overall mean added cost of $27,259. Added costs increased with the number of complications; isolated recurrent laryngeal nerve paresis ($3,911) and respiratory failure ($35,011) were associated with the least and most additive cost, respectively.

Conclusions: Lobectomy is associated with substantial variability of episode-of-care costs. Variability is driven by patient demographic and clinical factors, hospital characteristics, and the occurrence and severity of complications.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.athoracsur.2020.08.073DOI Listing
November 2020

Initial and Longitudinal Cost of Surgical Resection for Lung Cancer.

Ann Thorac Surg 2020 Oct 5. Epub 2020 Oct 5.

Department of Surgery, Emory University School of Medicine, Atlanta, Georgia.

Background: The longitudinal cost of treating patients with non-small cell lung cancer (NSCLC) undergoing surgical resection has not been evaluated. We describe initial and 4-year resource use and cost for NSCLC patients aged 65 years of age or greater who were treated surgically between 2008 and 2013.

Methods: Using clinical data for NSCLC resections from The Society of Thoracic Surgeons General Thoracic Surgery Database linked to Medicare claims, resource use and cost of preoperative staging, surgery, and subsequent care through 4 years were examined ($2017). Cost of hospital-based care was estimated using cost-to-charge ratios; professional services and care in other settings were valued using reimbursements. Inverse probability weighting was used to account for administrative censoring. Outcomes were stratified by pathologic stage and by surgical approach for stage I lobectomy patients.

Results: Resection hospitalizations averaged 6 days and cost $31,900. In the first 90 days, costs increased with stage ($12,430 for stage I to $26,350 for stage IV). Costs then declined toward quarterly means more similar among stages. Cumulative costs ranged from $131,032 (stage I) to $205,368 (stage IV). In the stage I lobectomy cohort, patients selected for minimally invasive procedures had lower 4-year costs than did thoracotomy patients ($120,346 versus $136,250).

Conclusions: The 4-year cost of surgical resection for NSCLC was substantial and increased with pathologic stage. Among stage I lobectomy patients, those selected for minimally invasive surgery had lower costs, particularly through 90 days. Potential avenues for improving the value of surgical resection include judicious use of postoperative intensive care and earlier detection and treatment of disease.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.athoracsur.2020.07.048DOI Listing
October 2020

The Evolving Burden of Drug Use Associated Infective Endocarditis in the United States.

Ann Thorac Surg 2020 10 6;110(4):1185-1192. Epub 2020 May 6.

Department of Cardiovascular Surgery, Mayo Clinic, Rochester, Minnesota. Electronic address:

Background: The rise in the number of valve operations performed for infective endocarditis (IE) due to drug use is an important manifestation of the opioid epidemic. This study characterized national trends and outcomes of valve surgery for drug use-associated IE (DU-IE).

Methods: Adults undergoing valve surgery for active IE in The Society of Thoracic Surgeons (STS) Adult Cardiac Surgery Database between July 2011 and June 2018 were stratified as DU-IE and non-DU-IE. Trends and clinical profiles were analyzed. Early outcomes were assessed. The association of DU-IE with outcomes was analyzed with multivariable regression, adjusting for STS Valve Risk model covariates.

Results: There were 34,905 valve operations performed for IE, of which 33.7% were for DU-IE. DU-IE operations increased 2.7-fold during the study period. There was considerable regional variability in DU-IE operations, ranging from 28% to 58% of all IE surgeries in 2018, with highest rates observed in East South Central and South Atlantic regions. DU-IE patients were younger and had fewer cardiovascular comorbidities. Risk-adjusted major morbidity and in-hospital mortality were significantly higher in the DU-IE group. Redo valve procedures in DU-IE patients were associated with worse outcomes, compared with those receiving a first valve operation.

Conclusions: Operations for DU-IE have increased sharply in the United States during the last several years, exhibiting substantial regional variability. DU-IE patients have unique clinical profiles, and worse risk-adjusted outcomes. This demonstrates the significant impact of the opioid epidemic on endocarditis surgeries and punctuates the urgent need for multidisciplinary regional and national efforts to reverse this trend.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.athoracsur.2020.03.089DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7502470PMC
October 2020

Associations between anthropometric indices and outcomes of congenital heart operations in infants and young children: An analysis of data from the Society of Thoracic Surgeons Database.

Am Heart J 2020 06 19;224:85-97. Epub 2020 Mar 19.

Anesthesiology and Pain Medicine, Seattle Children's Hospital, University of Washington, 4800 Sandpoint Way NE, Seattle, WA.

Background: Children with congenital heart disease are at risk for growth failure due to inadequate nutrient intake and increased metabolic demands. We examined the relationship between anthropometric indices of nutrition (height-for-age z-score [HAZ], weight-for-age z-score [WAZ], weight-for-height z-score [WHZ]) and outcomes in a large sample of children undergoing surgery for congenital heart disease.

Methods: Patients in the Society of Thoracic Surgeons Congenital Heart Surgery Database having index cardiac surgery at age 1 month to 10 years were included. Indices were calculated by comparing patients' weight and height to population norms from the World Health Organization and Centers for Disease Control and Prevention. Outcomes included operative mortality, composite mortality or major complication, major postoperative infection, and postoperative length of stay. For each outcome and index, the adjusted odds ratio (aOR) (for mortality, composite outcome, and infection) and adjusted relative change in median (for postoperative length of stay) for a 1-unit decrease in index were estimated using mixed-effects logistic and log-linear regression models.

Results: Every unit decrease in HAZ was associated with 1.40 aOR of mortality (95% CI 1.32-1.48), and every unit decrease in WAZ was associated with 1.33 aOR for mortality (95% CI 1.25-1.41). The relationship between WHZ and outcome was nonlinear, with aOR of mortality of 0.84 (95% CI 0.76-0.93) for 1-unit decrease when WHZ ≥ 0 and a nonsignificant association for WHZ < 0. Trends for other outcomes were similar. Overall, the incidence of low nutritional indices was similar for 1-ventricle and 2-ventricle patients. Children between the age of 1 month and 1 year and those with lesions associated with pulmonary overcirculation had the highest incidence of low nutritional indices.

Conclusions: Lower HAZ and WAZ, suggestive of malnutrition, are associated with increased mortality and other adverse outcomes after cardiac surgery in infants and young children. Higher WHZ over zero, suggestive of obesity, is also associated with adverse outcomes.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.ahj.2020.03.012DOI Listing
June 2020

Equivalent Survival Between Lobectomy and Segmentectomy for Clinical Stage IA Lung Cancer.

Ann Thorac Surg 2020 12 29;110(6):1882-1891. Epub 2020 Feb 29.

Division of Cardiothoracic Surgery, Emory University, Atlanta, Georgia.

Background: The oncologic efficacy of segmentectomy is controversial. We compared long-term survival in clinical stage IA (T1N0) Medicare patients undergoing lobectomy and segmentectomy in The Society of Thoracic Surgeons database.

Methods: The Society of Thoracic Surgeons General Thoracic Surgery Database was linked to Medicare data in 14,286 lung cancer patients who underwent segmentectomy (n = 1654) or lobectomy (n = 12,632) for clinical stage IA disease from 2002 to 2015. Cox regression was used to create a long-term survival model. Patients were then propensity matched on demographic and clinical variables to derive matched pairs.

Results: In Cox modeling segmentectomy was associated with survival similar to lobectomy in the entire cohort (hazard ratio, 1.04; 95% confidence interval, 0.89-1.20; P = .64) and in the matched subcohort. A subanalysis restricted to the 2009 to 2015 population (n = 11,811), when T1a tumors were specified and positron emission tomography results and mediastinal staging procedures were accurately recorded in the database, also showed that segmentectomy and lobectomy continue to have similar survival (hazard ratio, 1.00; 95% confidence interval, 0.87-1.16). Subanalysis of the pathologic N0 patients demonstrated the same results.

Conclusions: Lobectomy and segmentectomy for early-stage lung cancer are equally effective treatments with similar survival. Surgeons from The Society of Thoracic Surgeons database appear to be selecting patients appropriately for sublobar procedures.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.athoracsur.2020.01.020DOI Listing
December 2020

Analytical Treatment Interruption after Short-Term Antiretroviral Therapy in a Postnatally Simian-Human Immunodeficiency Virus-Infected Infant Rhesus Macaque Model.

mBio 2019 09 5;10(5). Epub 2019 Sep 5.

Duke Human Vaccine Institute, Duke University Medical Center, Durham, North Carolina, USA

To achieve long-term viral remission in human immunodeficiency virus (HIV)-infected children, novel strategies beyond early antiretroviral therapy (ART) will be necessary. Identifying clinical predictors of the time to viral rebound upon ART interruption will streamline the development of novel therapeutic strategies and accelerate their evaluation in clinical trials. However, identification of these biomarkers is logistically challenging in infants, due to sampling limitations and the potential risks of treatment interruption. To facilitate the identification of biomarkers predicting viral rebound, we have developed an infant rhesus macaque (RM) model of oral simian-human immunodeficiency virus (SHIV) SHIV.CH505.375H.dCT challenge and analytical treatment interruption (ATI) after short-term ART. We used this model to characterize SHIV replication kinetics and virus-specific immune responses during short-term ART or after ATI and demonstrated plasma viral rebound in 5 out of 6 (83%) infants. We observed a decline in humoral immune responses and partial dampening of systemic immune activation upon initiation of ART in these infants. Furthermore, we monitored SHIV replication and rebound kinetics in infant and adult RMs and found that both infants and adults demonstrated equally potent virus-specific humoral immune responses. Finally, we validated our models by confirming a well-established correlate of the time to viral rebound, namely, the pre-ART plasma viral load, as well as identified additional potential humoral immune correlates. Thus, this model of infant ART and viral rebound can be used and further optimized to define biomarkers of viral rebound following long-term ART as well as to preclinically assess novel therapies to achieve a pediatric HIV functional cure. Novel interventions that do not rely on daily adherence to ART are needed to achieve sustained viral remission for perinatally infected children, who currently rely on lifelong ART. Considering the risks and expense associated with ART interruption trials, the identification of biomarkers of viral rebound will prioritize promising therapeutic intervention strategies, including anti-HIV Env protein therapeutics. However, comprehensive studies to identify those biomarkers are logistically challenging in human infants, demanding the need for relevant nonhuman primate models of HIV rebound. In this study, we developed an infant RM model of oral infection with simian-human immunodeficiency virus expressing clade C HIV Env and short-term ART followed by ATI, longitudinally characterizing the immune responses to viral infection during ART and after ATI. Additionally, we compared this infant RM model to an analogous adult RM rebound model and identified virologic and immunologic correlates of the time to viral rebound after ATI.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1128/mBio.01971-19DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6945967PMC
September 2019

Lack of correlation between short- and long-term performance after lung cancer surgery.

J Thorac Cardiovasc Surg 2019 04 27;157(4):1633-1643.e3. Epub 2018 Nov 27.

Baptist M.D. Anderson Cancer Center, Jacksonville, Fla.

Objective: Outcomes for lung cancer surgery are currently measured according to perioperative morbidity and mortality. However, the oncologic efficacy of the surgery is reflected by long-term survival. We examined correlation between measures of short-term and long-term performance for lung cancer surgery.

Methods: The Society of Thoracic Surgeons General Thoracic Surgery Database linked to Medicare survival data was queried for pathologic stage I lung cancer resected between 2009 and 2013. Two separate multivariable models were created: (1) short-term: avoidance of perioperative major morbidity and mortality; and (2) long-term: 3-year survival. Standardized incidence ratios were calculated for the Society of Thoracic Surgeons programs (participants) to determine risk-adjusted participant performance measures for the short- and long-term time points. Correlation of participant standardized incidence ratios for short- and long-term performance was assessed using the Pearson correlation coefficient.

Results: The study population included 12,596 patients from 229 participating programs. One hundred fifty-one participants met minimum volume and follow-up requirements for analysis. Overall, performance for the short-term measure was uniform with only 2 (1.3%) participants performing better than expected and 2 (1.3%) worse than expected. For the long-term measure, 9 (6%) participants achieved better than expected and 5 (3.3%) worse than expected survival. No participant was an above or below average performer for the short- and long-term measures. Further, no correlation was observed between participant short- and long-term performance (Pearson correlation coefficient, 0.12; 95% confidence interval, -0.04 to 0.28; P = .14).

Conclusions: Avoidance of perioperative morbidity and mortality is an incomplete measure of performance in lung cancer surgery. Lung cancer surgery performance metrics should assess the safety of surgery and long-term survival.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.jtcvs.2018.09.141DOI Listing
April 2019

National Benchmarks for Proportions of Patients Receiving Blood Transfusions During Pediatric and Congenital Heart Surgery: An Analysis of the STS Congenital Heart Surgery Database.

Ann Thorac Surg 2018 10 11;106(4):1197-1203. Epub 2018 Jun 11.

Division of Cardiovascular Surgery, Johns Hopkins All Children's Heart Institute, Johns Hopkins All Children's Hospital, Saint Petersburg, Tampa, and Orlando, Florida; Division of Cardiac Surgery, Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland.

Background: To determine national benchmarks and assess variability across centers, The Society of Thoracic Surgeons Congenital Heart Surgery Database was analyzed to document proportions of patients receiving intraoperative transfusion of packed red blood cells (PRBC) during open heart surgery.

Methods: Index cardiopulmonary bypass operations reported in The Society of Thoracic Surgeons Congenital Heart Surgery Database (2014 to 2015) were potentially eligible for inclusion. Data from centers with more than 15% missing data for PRBC transfusion were excluded, as were individual records missing information about PRBC transfusion. The distribution of center-level PRBC transfusion rates in various clinically relevant groups was estimated by fitting a two-level logistic mixed model.

Results: The study population included 22,874 index cardiopulmonary bypass operations in 81 centers. Center-level intraoperative PRBC transfusion rates stratified by age group, weight, STAT Mortality Category, and lowest core temperature were documented. For younger patients and patients undergoing higher-complexity operations, median center PRBC transfusion rates consistently approached 100%, with narrow interquartile ranges indicating little center variability. Center PRBC transfusion rates declined with increasing patient age, but with greater variability (wider interquartile ranges) across centers. Intraoperative PRBC transfusion was uncommon (median center transfusion rates <30%) in older patients (teenagers and adults) undergoing lower-complexity (STAT Mortality Category <3) operations.

Conclusions: Most centers transfuse PRBCs routinely in higher-risk, younger, and smaller patients, with little variability across centers. For lower-risk operations in older and larger patients, centers are more likely to forgo intraoperative transfusions. This analysis provides national benchmarks for center-level PRBC transfusion rates during pediatric and congenital heart surgery.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.athoracsur.2018.04.088DOI Listing
October 2018

The Society of Thoracic Surgeons 2018 Adult Cardiac Surgery Risk Models: Part 1-Background, Design Considerations, and Model Development.

Ann Thorac Surg 2018 05 22;105(5):1411-1418. Epub 2018 Mar 22.

Duke Clinical Research Institute, Duke University Medical Center, Durham, North Carolina.

Background: The last published version of The Society of Thoracic Surgeons (STS) Adult Cardiac Surgery Database (ACSD) risk models were developed in 2008 based on patient data from 2002 to 2006 and have been periodically recalibrated. In response to evolving changes in patient characteristics, risk profiles, surgical practice, and outcomes, the STS has now developed a set of entirely new risk models for adult cardiac surgery.

Methods: New models were estimated for isolated coronary artery bypass grafting surgery (CABG [n = 439,092]), isolated aortic or mitral valve surgery (n = 150,150), and combined valve plus CABG procedures (n = 81,588). The development set was based on July 2011 to June 2014 STS ACSD data; validation was performed using July 2014 to December 2016 data. Separate models were developed for operative mortality, stroke, renal failure, prolonged ventilation, reoperation, composite major morbidity or mortality, and prolonged or short postoperative length of stay. Because of its low occurrence rate, a combined model incorporating all operative types was developed for deep sternal wound infection/mediastinitis.

Results: Calibration was excellent except for the deep sternal wound infection/mediastinitis model, which slightly underestimated risk because of higher rates of this endpoint in the more recent validation data; this will be recalibrated in each feedback report. Discrimination (c-index) of all models was superior to that of 2008 models except for the stroke model for valve patients.

Conclusions: Completely new STS ACSD risk models have been developed based on contemporary patient data; their performance is superior to that of previous STS ACSD models.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.athoracsur.2018.03.002DOI Listing
May 2018

The Society of Thoracic Surgeons 2018 Adult Cardiac Surgery Risk Models: Part 2-Statistical Methods and Results.

Ann Thorac Surg 2018 05 22;105(5):1419-1428. Epub 2018 Mar 22.

Department of Surgery and Center for Quality and Safety, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts. Electronic address:

Background: The Society of Thoracic Surgeons (STS) uses statistical models to create risk-adjusted performance metrics for Adult Cardiac Surgery Database (ACSD) participants. Because of temporal changes in patient characteristics and outcomes, evolution of surgical practice, and additional risk factors available in recent ACSD versions, completely new risk models have been developed.

Methods: Using July 2011 to June 2014 ACSD data, risk models were developed for operative mortality, stroke, renal failure, prolonged ventilation, mediastinitis/deep sternal wound infection, reoperation, major morbidity or mortality composite, prolonged postoperative length of stay, and short postoperative length of stay among patients who underwent isolated coronary artery bypass grafting surgery (n = 439,092), aortic or mitral valve surgery (n = 150,150), or combined valve plus coronary artery bypass grafting surgery (n = 81,588). Separate models were developed for each procedure and endpoint except mediastinitis/deep sternal wound infection, which was analyzed in a combined model because of its infrequency. A surgeon panel selected predictors by assessing model performance and clinical face validity of full and progressively more parsimonious models. The ACSD data (July 2014 to December 2016) were used to assess model calibration and to compare discrimination with previous STS risk models.

Results: Calibration in the validation sample was excellent for all models except mediastinitis/deep sternal wound infection, which slightly underestimated risk and will be recalibrated in feedback reports. The c-indices of new models exceeded those of the last published STS models for all populations and endpoints except stroke in valve patients.

Conclusions: New STS ACSD risk models have generally excellent calibration and discrimination and are well suited for risk adjustment of STS performance metrics.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.athoracsur.2018.03.003DOI Listing
May 2018

Comparing outcomes of early, late, and non-surgical management of intraspinal abscess.

J Clin Neurosci 2017 Feb 9;36:64-71. Epub 2016 Nov 9.

Department of Neurosurgery, Duke University Medical Center, Durham, NC 27710, United States. Electronic address:

Intraspinal abscesses (ISAs) are rare lesions that are often neurologically devastating. Current treatment paradigms vary widely including early surgical decompression, drainage, and systemic antibiotics, delayed surgery, and sole medical management. The National Inpatient Sample (NIS) database was queried for cases of ISA from 2003 to 2012. Early and late surgery were defined as occurring before or after 48h of admission. Outcome measures included mortality, incidence of major complications, length of stay (LOS), and inpatient costs. A total of 10,150 patients were included (6281 early surgery, 3167 delayed surgery, 702 medical management). Paralysis, the main comorbidity, was most associated with early surgery (p<0.0001). In multivariate analysis, the rates of postoperative infection and paraplegia were highest with early surgery (p<0.0001), but the incidence of sepsis was higher with delayed surgery (p<0.0001). Early surgery was least associated with in-hospital mortality (p=0.0212), sepsis (p<0.001), and had the shortest LOS (p<0.001). Charges were highest with delayed surgery, and least with medical management (p<0.001). Medical management was associated with lower rates of complications (p<0.001). This is the largest study of patients with ISAs ever performed. Our results suggest that patients with ISAs undergoing surgical management have better outcomes and lower costs when operated on within 48h of admission, emphasizing the importance of accurate and early diagnosis of ISA.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.jocn.2016.10.035DOI Listing
February 2017