Objective The study objective was to compare the outcomes of intraoperative extracorporeal membrane oxygenation versus cardiopulmonary bypass support in lung transplantation. Methods We performed a ...retrospective cohort study from a prospective database of adult lung transplantations performed at the University of Toronto from 2007 to 2013. Among 673 lung transplantations performed in the study period, 267 (39.7%) required cardiopulmonary support. There were 39 cases of extracorporeal membrane oxygenation (2012-2013) and 228 cases of cardiopulmonary bypass (2007-2013). Patients who were bridged with extracorporeal life support, underwent a concomitant cardiac procedure, received a combined liver or heart transplant, were colonized with Burkholderia cenocepacia , or required emergency cannulation for cardiopulmonary support were excluded. Finally, 33 extracorporeal membrane oxygenation cases were matched with 66 cases of cardiopulmonary bypass according to age (±10 years), lung transplantation indication, and procedure type (bilateral vs single lung transplantation). Results Recipient factors such as body mass index and gender were not different between extracorporeal membrane oxygenation and cardiopulmonary bypass groups. Furthermore, donor variables were similar, including age, body mass index, last PaO2/FiO2 ratio, smoking history, positive airway cultures, and donor type (brain death and donation after cardiac death). Early outcomes, such as mechanical ventilation requirement, length of intensive care unit stay, and length of hospital stay, significantly favored extracorporeal membrane oxygenation (median 3 vs 7.5 days, P = .005; 5 vs 9.5 days, P = .026; 19 vs 27 days, P = .029, respectively). Perioperative blood product transfusion requirement was lower in the extracorporeal membrane oxygenation group. The 90-day mortality for the extracorporeal membrane oxygenation group was 6% versus 15% for cardiopulmonary bypass ( P = .32). Conclusions Extracorporeal membrane oxygenation may be considered as the first choice of intraoperative cardiorespiratory support for lung transplantation.
Abstract Objective(s) Localization and resection of non-visible, non-palpable pulmonary nodules during video-assisted thoracoscopic surgery (VATS) is challenging. Our study was to determine the ...feasibility and safety of indocyanine green (ICG) fluorescence localization and resection of small nodules using a near-infrared (NIR) fluorescence thoracoscope. Methods Twenty patients with undiagnosed peripheral nodules smaller than 3cm scheduled for CT-guided microcoil placement followed by VATS wedge resection were enrolled. After microcoil deployment, 100-150 μl of diluted ICG was injected percutaneously near the nodule. The nodule was initially localized solely by using the NIR thoracoscope to visualize ICG fluorescence. Thoracoscopic instruments were used to determine the staple line. Wedge resection was performed after confirmation of the location of the microcoil using fluoroscopy. Results Twenty patients underwent NIR image-guided VATS resection. The median CT tumor size was 1.2 cm. The median depth from the pleural surface was 1.4 cm (range: 0.2-4.8). The median CT-guided intervention time was 35 min and VATS procedural time was 54 min. ICG fluorescence was clearly identified in 18 of 20 cases (90%). The surgical margins were all negative on final pathology without the need of additional resection. The final diagnoses included 18 primary lung cancer, 1 metastatic lung cancer, and 1 benign lung tumor. Conclusions CT-guided percutaneous ICG injection and intraoperative NIR localization of small nodules is safe and feasible. It offers surgeons the ease of localization through direct ICG fluorescence imaging without the use of fluoroscopy and may be a complementary technique to preoperative microcoil placement for non-visible, non-palpable intrapulmonary nodules.
Objective Normothermic ex vivo lung perfusion is a novel method to evaluate and improve the function of injured donor lungs. We reviewed our experience with 50 consecutive transplants after ex vivo ...lung perfusion. Methods A retrospective study using prospectively collected data was performed. High-risk brain death donor lungs (defined as Pa o2 /F io2 <300 mm Hg or lungs with radiographic or clinical findings of pulmonary edema) and lungs from cardiac death donors were subjected to 4 to 6 hours of ex vivo lung perfusion. Lungs that achieved stable airway and vascular pressures and Pa o2 /F io2 greater than 400 mm Hg during ex vivo lung perfusion were transplanted. The primary end point was the incidence of primary graft dysfunction grade 3 at 72 hours after transplantation. End points were compared with lung transplants not treated with ex vivo lung perfusion (controls). Results A total of 317 lung transplants were performed during the study period (39 months). Fifty-eight ex vivo lung perfusion procedures were performed, resulting in 50 transplants (86% use). Of these, 22 were from cardiac death donors and 28 were from brain death donors. The mean donor Pa o2 /F io2 was 334 mm Hg in the ex vivo lung perfusion group and 452 mm Hg in the control group ( P = .0001). The incidence of primary graft dysfunction grade 3 at 72 hours was 2% in the ex vivo lung perfusion group and 8.5% in the control group ( P = .14). One patient (2%) in the ex vivo lung perfusion group and 7 patients (2.7%) in the control group required extracorporeal lung support for primary graft dysfunction ( P = 1.00). The median time to extubation, intensive care unit stay, and hospital length of stay were 2, 4, and 20 days, respectively, in the ex vivo lung perfusion group and 2, 4, and 23 days, respectively, in the control group ( P > .05). Thirty-day mortality (4% in the ex vivo lung perfusion group and 3.5% in the control group, P = 1.00) and 1-year survival (87% in the ex vivo lung perfusion group and 86% in the control group, P = 1.00) were similar in both groups. Conclusions Transplantation of high-risk donor lungs after 4 to 6 hours of ex vivo lung perfusion is safe, and outcomes are similar to those of conventional transplants. Ex vivo lung perfusion improved our center use of donor lungs, accounting for 20% of our current lung transplant activity.
Abstract Objective To assess the cost-effectiveness of various modes of mediastinal staging in non–small cell lung cancer (NSCLC) in a single-payer health care system. Methods We performed a decision ...analysis to compare the health outcomes and costs of 4 mediastinal staging strategies: no invasive staging, endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA), mediastinoscopy, and EBUS-TBNA followed by mediastinoscopy if EBUS-TBNA is negative. We determined incremental cost effectiveness ratios (ICER) for all strategies and performed comprehensive deterministic sensitivity analyses using a willingness to pay threshold of $80,000/quality adjusted life year (QALY). Results Under the base-case scenario, the no invasive mediastinal staging strategy was least effective (QALY, 5.80) and least expensive ($11,863), followed by mediastinoscopy, EBUS-TBNA, and EBUS-TBNA followed by mediastinoscopy with 5.86, 5.87, and 5.88 QALYs, respectively. The ICER was ∼$26,000/QALY for EBUS-TBNA staging and ∼$1,400,000/QALY for EBUS-TBNA followed by mediastinoscopy. The mediastinoscopy strategy was dominated. Once pN2 exceeds 2.5%, EBUS-TBNA staging is cost-effective (∼$80,000/QALY). Once the pN2 reaches 57%, EBUS-TBNA followed by mediastinoscopy is cost-effective (ICER ∼$79,000/QALY). Once EBUS-TBNA sensitivity exceeds 25%, EBUS-TBNA staging is cost-effective (ICER ∼$79,000/QALY). Once pN2 exceeds 25%, confirmatory mediastinoscopy should be added, in cases of EBUS-TBNA sensitivity ≤ 60%. Conclusions Invasive mediastinal staging in NSCLC is unlikely to be cost-effective in clinical N0 patients if pN2 <2.5%. In patients with probability of mediastinal metastasis between 2.5% and 57% EBUS-TBNA is cost-effective as the only staging modality. Confirmatory mediastinoscopy should be considered in high-risk patients (pN2 > 57%) in case of negative EBUS-TBNA.
Objective The study objective was to compare endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) with mediastinoscopy for mediastinal lymph node staging of potentially ...resectable non–small cell lung cancer. Methods Patients with confirmed or suspected non–small cell lung cancer who required mediastinoscopy to determine suitability for lung cancer resection were entered into the trial. All patients underwent EBUS-TBNA followed by mediastinoscopy under general anesthesia. If both were negative for N2 or N3 disease, the patient underwent pulmonary resection and mediastinal lymphadenectomy. Results Between July 2006 and August 2010, 190 patients were registered in the study, 159 enrolled, and 153 were eligible for analysis. EBUS-TBNA and mediastinoscopy sampled an average of 3 and 4 lymph node stations per patient, respectively. The mean short axis of the lymph node biopsied by EBUS-TBNA was 6.9 ± 2.9 mm. The prevalence of N2/N3 disease was 35% (53/153). There was excellent agreement between EBUS-TBNA and mediastinoscopy for mediastinal staging in 136 patients (91%; Kappa, 0.8; 95% confidence interval, 0.7–0.9). Specificity and positive predictive value for both techniques were 100%. The sensitivity, negative predictive value, and diagnostic accuracy for mediastinal lymph node staging for EBUS-TBNA and mediastinoscopy were 81%, 91%, 93%, and 79%, 90%, 93%, respectively. No significant differences were found between EBUS-TBNA and mediastinoscopy in determining the true pathologic N stage (McNemar’s test, P = .78). There were no complications from EBUS-TBNA. Minor complications from mediastinoscopy were observed in 4 patients (2.6%). Conclusions EBUS-TBNA and mediastinoscopy achieve similar results for the mediastinal staging of lung cancer. As performed in this study, EBUS-TBNA can replace mediastinoscopy in patients with potentially resectable non–small cell lung cancer.
The translation of novel drugs in lung transplantation is challenged by different physiologic conditions between small animals and humans. Large-animal models provide important pre-clinical evidence ...and the next step that best informs clinical trials. In the present study, we used a pig lung transplant model to determine whether human α1-antitrypsin (A1AT), a medication shown to prevent pulmonary ischemia-reperfusion injury in rats, could attenuate reperfusion injury after prolonged hypothermic preservation in a large-animal lung transplant model.
Donor lungs were preserved for 24 hours at 4°C, followed by lung transplantation. In a randomized and blinded fashion, intravenous A1AT (240 mg/kg; n = 5) or human albumin (n = 5) was administered to the recipient before reperfusion. Allograft gas exchange function and lung mechanics were monitored during a 4-hour reperfusion period. Microscopic lung injury, inflammatory response, coagulation activity, and cell death were assessed.
Pulmonary gas exchange was significantly better during the 4-hour reperfusion period in the A1AT group. Treatment with A1AT improved static pulmonary compliance and significantly reduced pulmonary edema and lung permeability. A1AT treatment inhibited inflammatory mediators in the circulation, with reduced activation of nuclear factor-κB and inflammasome, reduced formation of thrombin-antithrombin complex in plasma, and reduced apoptosis in the allografts.
Administration of human A1AT before reperfusion in recipients improved immediate post-transplant lung function in pigs. A large-animal survival model should be considered to support further advancement toward a clinical trial of A1AT to prevent primary graft dysfunction in lung transplantation.
Objectives We and others have reported the early experience with off-label use of thoracic aortic endografts to facilitate the resection of tumors infiltrating the aorta. We describe our extended ...experience and long-term outcome using this innovative approach. Methods Patients with preoperative suspected thoracic aortic infiltration who underwent endografting followed by en bloc tumor resection including the aortic wall were retrospectively reviewed and data were analyzed. Results Between 2008 and 2012, 5 patients (4 female) with a median age of 52 years (34-63 years) were included. Tumors infiltrating the aorta were non–small cell lung carcinomas (n = 3) and sarcomas (n = 2). Both patients with sarcoma had neoadjuvant radiation, whereas patients with non–small cell lung carcinomas had neoadjuvant (n = 2) or adjuvant chemoradiation (n = 1). Aortic endografting was performed 1 to 17 days before resection of the tumor. The proximal end of the stent-graft was deployed in the aortic arch (n = 2) or the descending aorta (n = 3). The tumor was resected en bloc in all patients and combined with chest wall and 2 to 3 levels of spinal resection in 4 of the 5 patients. Two patients with full-thickness aortic wall resection had additional buttressing of the defect. Cardiopulmonary bypass was never required. One patient had an empyema requiring debridements and thoracic window. After a median follow-up of 39 months (range, 9-62 months), all patients were alive and disease-free. None of them had overt endograft-related complications. Conclusions Thoracic aortic endografting allowed safe en bloc resection of tumors invading the aortic wall. Therefore, the indication for thoracic aortic endografts could be extended to specific oncologic cases.
Objectives A minimal-dose computed tomography scan of the thorax (MnDCT) delivers a radiation dose comparable with a chest x-ray (CXR). We hypothesized that in patients with completely resected lung ...cancer, surveillance with MnDCT, when compared with CXR, leads to earlier detection and a higher rate of treatment of new or recurrent lung cancer. Methods After lung cancer resection, patients prospectively were enrolled for surveillance with MnDCT and CXR at 3, 6, 12, 18, 24, 36, 48, and 60 months. Images were interpreted by different blinded radiologists. When new or recurrent cancer was suspected, standard-dose CT and/or a tissue biopsy were performed for confirmation. Results Between 2007 and 2012, 271 patients were included and 1137 pairs of CXR and MnDCT were analyzed. MnDCT was more sensitive (94% vs 21%; P < .0001) and had a higher negative predictive value (99% vs 96%; P = .007) than CXR for the diagnosis of new or recurrent lung cancer. The prevalence of new or recurrent lung cancer was 23.2% (63 of 271), of whom 78% (49 of 63) had asymptomatic disease. The majority of asymptomatic patients (75%; 37 of 49) were treated with curative intent and had a median survival of 69 months. The remainder of patients received palliative treatment (24%; 12 of 49) and had a median survival of 25 months ( P < .0001). Conclusions After curative resection of lung cancer, MnDCT is superior to CXR for the detection of new or recurrent lung cancer. The majority of new or recurrent cancer was detected by MnDCT at an asymptomatic phase, allowing for curative treatment, leading to a long survival.
Objectives In vivo lung perfusion (IVLP) is an emergent strategy to treat lung metastases because it allows localized delivery of chemotherapy with minimal systemic exposure. Previously, short-term ...(±30 minutes) IVLP resulted in variable efficacy and significant lung toxicity. We hypothesize that a modified IVLP strategy derived from an ex vivo lung perfusion technique could minimize lung injury. Our objective was to demonstrate the feasibility and safety of a modified prolonged (4 hours) IVLP. Methods Six Yorkshire pigs were used for the experiments. A thoracotomy was performed, the left pulmonary artery and pulmonary veins were cannulated, and the left lung was isolated in situ. IVLP was performed at normothermia for 4 hours using Steen Solution (XVIVO Perfusion, Göteburg, Sweden) as perfusate. The flow rate was 16% of estimated cardiac output and left atrial pressure was maintained between 3 and 5 mm Hg. Perfusate was deoxygenated and supplied with CO2 to physiologic levels before entering the lungs. A protective mode of ventilation was used. After IVLP, the left lung was allowed to reperfuse for additional 4 hours. Airway dynamics, gas exchange, and pulmonary vascular resistance were used to assess left lung physiology. Histologic signs of lung injury were assessed before and after IVLP, and 4 hours after reperfusion. Results Lung function parameters were stable throughout the 4-hour IVLP and during reperfusion. No significant histologic evidence of acute lung injury was observed. Conclusions Four hours of IVLP is feasible without adding significant lung injury. Prolonged perfusion time and a protective protocol might provide safer and more efficacious treatment of pulmonary metastases.
Abstract Objective We hypothesized that administration of a homodimer of recombinant annexin V, diannexin, could shield phosphatidylserine on the endothelium, and inhibit leukocyte and platelet ...adhesion, thereby potentially reducing ischemia reperfusion injury (IRI) in lung transplantation. This hypothesis was tested using a rat syngeneic single left-lung transplant model. Methods Rats were randomly assigned to receive diannexin (DN group; n = 10) or normal saline (control group; n = 10). Diannexin (1000 μg/kg) was administered to the donor lung in the pulmonary flush solution, and to the recipient intravenously, 5 minutes after initiation of reperfusion. Grafts were reperfused for 2 hours. Results The transplanted grafts in the DN group performed significantly better in gas exchange with higher partial pressure of oxygen (control group: 179 ± 121 vs DN group: 330 ± 54 mm Hg; P = .007) and lower partial pressure of carbon dioxide (control: 55.1 ± 26 vs DN: 34.2 ± 11 mm Hg; P = .04), as well as lower peak airway pressure (control: 20.5 ± 8.5 vs DN: 12.0 ± 7.9 cm H2 O; P = .035) after 2 hours of reperfusion. Wet-to-dry lung weight ratio ( P = .054), and alveolar fibrin deposition score ( P = .04), were reduced in the DN group. Caspase-cleaved cytokeratin 18 in plasma (a marker of epithelial apoptosis) was significantly reduced in the DN group ( P = .013). Furthermore, gene-expression levels of proinflammatory cytokines in the transplanted graft, including interleukin-6 ( P = .04) and macrophage inflammatory protein 2 ( P = .03) were significantly decreased in the DN group. Conclusions A homodimer of recombinant annexin V reduced ischemia reperfusion injury in a lung transplant animal model, by reducing cell death and tissue inflammation.
PDF
