Annals of Thoracic Surgery最新文献

Background: Allografts from donation after circulatory death (DCD) or brain death donors may be evaluated by ex vivo lung perfusion (EVLP) to assess quality for transplantation. We sought to determine the association of donor type with transplantation outcomes at a national level.

Methods: The United Network for Organ Sharing database was queried for lung transplant recipients, which were stratified into: DCD EVLP, brain death EVLP, standard DCD and standard brain death, followed by an unadjusted analysis. 1:1 propensity matching based on donor and recipient characteristics was used to compare DCD v DCD EVLP, brain death v brain death EVLP and brain death v DCD EVLP. The cohorts were assessed with comparative statistics. Finally, static and portable EVLP were compared to determine independent association with increased mortality.

Results: The unadjusted DCD EVLP group had significantly higher incidence of post-operative morbidity and mortality. 3-year survival was significantly lower in the DCD EVLP group, 65.3% (p=0.026). Following matching, the EVLP groups had significantly higher morbidity, and in-hospital mortality (DCD EVLP v brain death), but mid-term survival was no longer significantly different. However, the DCD EVLP group had about ∼6% lower survival than the DCD group (p=0.05) and about ∼7% lower survival than the brain death group (p=0.12). Within the EVLP groups, static and portable EVLP were not independently associated with increased mortality.

Conclusions: Expansion of DCD EVLP allografts increases organ access, though providers should be aware of potential increases in complications and mortality as compared to DCD alone.

Background: There is concern that left internal thoracic artery (LITA) to diagonal to left anterior descending artery (LAD) grafts may be more susceptible to failure compared to single LITA-LAD grafts.

Methods: Pooled individual patient data from eight clinical trials with systematic graft imaging were analyzed to assess the incidence of sequential LITA-Diagonal-LAD vs. single LITA-LAD grafts. Mixed-effects multivariable logistic regression, adjusting for patient characteristics and clustering within trials, was used.

Results: Of 3969 patients with LITA-LAD grafts, 283 (7.1%) patients received sequential LITA-Diagonal-LAD grafts. Patients with sequential LITA-Diagonal-LAD grafts were older (66 vs. 65 y, p=0.009) and more often male (88% vs. 83%, p=0.03). Overall, graft failure occurred in 9.3% of patients with LITA-LAD grafts, with more graft failure occurring in single (9.5%) than in sequential LITA-Diagonal-LAD grafts (6.4%, p=0.08) at a median (25^th-75^th percentile) time to imaging of 1.0 (1.0-1.1) years. After multivariable adjustment, sequential LITA-Diagonal-LAD grafting was not associated with graft failure (adjusted odds ratio: 1.22, 95% confidence interval: 0.68-2.18, p=0.55). There was no difference in mortality (2.8% vs. 5.3%, p=0.06), myocardial infarction (1.4% vs. 1.6%, p=0.90), revascularization (4.5% vs. 7.3%, p=0.08), or stroke (1.7% vs. 1.2%, p=0.40) between groups.

Conclusions: In selected patients, LITA-Diagonal-LAD grafting was not associated with higher risk of graft failure or adverse clinical events at one year.

Background: Recent randomized trials have shown equivalent survival after sublobar resection (SLR) versus lobectomy in patients with clinical stage IA non-small cell lung cancer (NSCLC)≤2cm. High SUVmax is a known risk factor in NSCLC, yet limited data exists on whether a high SUV should preclude a SLR. This study aims to determine if there is an association between SUVmax and survival based on the extent of parenchymal resection.

Methods: A retrospective review of a prospectively maintained institutional database was conducted to identify patients with clinical stage IA NSCLC≤2cm (2011-2020) treated with SLR or lobectomy. The primary outcome was cancer-specific survival (CSS). Secondary outcomes were overall survival (OS) and disease-free survival (DFS).

Results: 543 patients were identified; 36.8% had SLR and 63.2% had lobectomy. Baseline characteristics were similar. Patients who had SLR had significantly worse ECOG performance status and higher rates of comorbidities. 5-year CSS, OS, and DFS for the whole cohort were similar between SLR and lobectomy. A receiver operating characteristic curve estimated the SUVmax cutoff point to be 4.15. For the whole cohort, patients with SUVmax>4.15 had worse CSS compared to SUVmax≤4.15. However, there was no significant difference in 5-year CSS after SLR versus lobectomy in patients with SUVmax≤4.15 (98% in both groups; P=0.77) or patients with SUVmax>4.15 (90% versus 94% respectively; P=0.12).

Conclusions: SUVmax may not be a useful clinical determinant of the extent of parenchymal resection in patients with cT1N0 NSCLC≤2cm. Patients treated by SLR had comparable survival to lobectomy, irrespective of PET avidity.

Background: Robotic approaches have been increasingly utilized for cardiothoracic operations, though concerns regarding costs remain. We evaluated short-term outcomes and costs of robotic-assisted and conventional mitral valve repair (MV-repair), hypothesizing that cost differences would be mitigated at high-volume programs.

Methods: Adults undergoing elective MV-repair from 2016 to 2020 were identified in the Nationwide Readmissions Database. Patients with rheumatic heart disease, mitral stenosis, and those undergoing concomitant operations were excluded. Generalized linear models were utilized to evaluate the association between approach and in-hospital mortality, complications, length of stay, costs, and 90-day readmissions. Annual institutional MV-repair volume was modeled using restricted cubic splines, and cost differences subsequently evaluated by volume tertile.

Results: Of 40,738 patients, 9.8% underwent robotic-assisted MV-repair. Risk-adjusted outcomes including mortality, stroke, reoperation, respiratory complications, postoperative infection, and readmission were comparable between the two groups, while those undergoing robotic-assisted MV-repair had lower rates of non-home discharge. The median cost of robotic-assisted MV-repair was greater than conventional surgery ($46,800 vs $38,500, p<0.001). Despite a 1.3-day decrement (95% CI 1.1-1.6) in length of stay, robotic-assisted MV-repair was associated with greater risk-adjusted costs by $10,500 (95% CI 5,800-15,200). Programs in the highest volume tertile exhibited comparable costs for robotic-assisted and conventional MV-repair (cost difference $5,900, 95% CI -1,200-12,200, p>0.05).

Conclusions: Robotic-assisted MV-repair had comparable short-term outcomes relative to conventional surgery. Despite increased costs of robotic-assisted MV-repair overall, high-volume programs had similar risk-adjusted costs by approach. These findings support the designation and performance of robotic MV-repair at centers of excellence in the United States.

Background: Patients may receive their adjuvant therapy at a facility different than where they had their lung cancer operation. Whether this fragmentation of care affects outcomes is unclear.

Methods: We used the National Cancer Database to identify lung cancer patients undergoing resection and adjuvant chemotherapy from 2006-2020. We stratified patients into those receiving fragmented care or not, and further divided fragmented care patients by the Commission on Cancer (CoC) accreditation status of the hospital. Fragmented care refers to patients receiving surgery and chemotherapy at different institutions. These institutions can be either CoC accredited or not. The main outcome was overall survival. We used Kaplan-Meier analysis to estimate survival and multivariable and Cox proportional models to identify associations.

Results: Of 65,369 patients, 32,494(49.7%) had fragmented care, with the majority(70.4%) receiving their chemotherapy at a non-CoC accredited facility. Factors associated with fragmented care were white(adjusted odds ratio(aOR)=1.34;p<0.001), lower comorbidity index(aOR=1.11;p<0.001), having a private insurance(aOR=1.11;p<0.001), and a higher median income(aOR=1.24;p<0.001). Fragmented care was associated with worse overall survival(Median survival=60vs65 months;p<0.001) compared to single center care. When care was fragmented, receiving adjuvant chemotherapy at CoC accredited centers had higher 5-year overall survival rates compared to those fragmented care at non-CoC centers(Median survival=71vs55 months;p<0.001).

Conclusions: The majority of lung cancer patients have their care fragmented to non-CoC accredited centers and this is associated with worse outcomes. Regionalization, achieving CoC accreditation, or improved patient access may be necessary to allow select patients to receive closer care while maintaining outcomes.

Background: Cardiothoracic surgery lacks gender and racial/ethnic diversity. Recent studies highlighted disparities based on gender and race/ethnicity among academic cardiothoracic surgeons. The impact of the intersection of these factors on representation and salary is unknown.

Methods: A cross-sectional analysis of Accreditation Council for Graduate Medical Education and Association of American Medical Colleges data was performed on the number of trainees and clinical faculty stratified by race/ethnicity and gender using Chi-square testing.

Results: The number of women and underrepresented minorities was low in cardiothoracic surgery compared to other specialties, with lowest representation at the intersection of race/ethnicity and gender. Among trainees, 8% were Asian, 2% were Black/African American , and 1.5% were Hispanic/Latina women. Among cardiothoracic faculty, 3.4% were Asian, 0.8% were Black/African American, and 0.4% were Hispanic/Latina women. Women in academic medicine, surgery and cardiothoracic surgery earned 80-87% the salary of men of equal academic rank. White assistant professors earned more than their colleagues (all clinical faculty, surgeons and cardiothoracic surgeons), this difference was further compounded by gender.

Conclusions: Salary disparities exist among cardiothoracic surgeons at the intersection of gender and race/ethnicity. Women experience salary disparity across all academic ranks and specialties. When considering the intersection of gender and race/ethnicity, gender is the predominant factor driving salary inequity.

Harvey W. Bender Jr spent his early years in Humble, Texas. After attending Baylor University College of Medicine, he trained in surgery at the Johns Hopkins Hospital. In 1971 he was recruited to Vanderbilt University to reinvigorate the residency training program and significantly expand the clinical services. He became Chair of the Residency Review Committee for Thoracic Surgery and of the American Board of Thoracic Surgery. He was also Chair of the Board of Regents, President of the American College of Surgeons, and President of the Southern Thoracic Surgical Association. He was a unique person whose influence will last for generations.