A Pilot Study Using Machine Learning Algorithms and Wearable Technology for the Early Detection of Postoperative Complications After Cardiothoracic Surgery.

OBJECTIVE: To evaluate whether a machine learning algorithm (i.e. the "NightSignal" algorithm) can be used for the detection of postoperative complications prior to symptom onset after cardiothoracic surgery. SUMMARY BACKGROUND DATA: Methods that enable the early detection of postoperative complications after cardiothoracic surgery are needed. METHODS: This was a prospective observational cohort study conducted from July 2021 to February 2023 at a single academic tertiary care hospital. Patients aged 18 years or older scheduled to undergo cardiothoracic surgery were recruited. Study participants wore a Fitbit watch continuously for at least 1 week preoperatively and up to 90-days postoperatively. The ability of the NightSignal algorithm-which was previously developed for the early detection of Covid-19-to detect postoperative complications was evaluated. The primary outcomes were algorithm sensitivity and specificity for postoperative event detection. RESULTS: A total of 56 patients undergoing cardiothoracic surgery met inclusion criteria, of which 24 (42.9%) underwent thoracic operations and 32 (57.1%) underwent cardiac operations. The median age was 62 (IQR: 51-68) years and 30 (53.6%) patients were female. The NightSignal algorithm detected 17 of the 21 postoperative events a median of 2 (IQR: 1-3) days prior to symptom onset, representing a sensitivity of 81%. The specificity, negative predictive value, and positive predictive value of the algorithm for the detection of postoperative events were 75%, 97%, and 28%, respectively. CONCLUSIONS: Machine learning analysis of biometric data collected from wearable devices has the potential to detect postoperative complications-prior to symptom onset-after cardiothoracic surgery.

Efficacy and Safety of Supraclavicular Thoracic Outlet Decompression.

OBJECTIVES: We aimed to report efficacy, safety, and health-related quality of life (HRQoL) outcomes of a multidisciplinary treatment approach including supraclavicular thoracic outlet decompression among patients with thoracic outlet syndrome (TOS). BACKGROUND: TOS is a challenging condition where controversy remains in diagnosis and treatment, primarily given a lack of data exploring various treatment approaches and associated patient outcomes. METHODS: Patients who underwent unilateral, supraclavicular thoracic outlet decompression, or pectoralis minor tenotomy for neurogenic, venous, or arterial TOS were identified from a prospectively maintained database. Demography, use of preoperative botulinum toxin injection, and participation in multidisciplinary evaluation were measured. The primary endpoints were composite postoperative morbidity and symptomatic improvement compared with baseline. RESULTS: Among 2869 patients evaluated (2007-2021), 1032 underwent surgery, including 864 (83.7%) supraclavicular decompressions and 168 (16.3%) isolated pectoralis minor tenotomies. Predominant TOS subtypes among surgical patients were neurogenic (75.4%) and venous TOS (23.4%). Most patients (92.9%) with nTOS underwent preoperative botulinum toxin injection; 56.3% reported symptomatic improvement. Before surgical consultation, few patients reported participation in physical therapy (10.9%). The median time from first evaluation to surgery was 136 days (interquartile range: 55, 258). Among 864 patients who underwent supraclavicular thoracic outlet decompression, complications occurred in 19.8%; the most common complication was chyle leak (8.3%). Four patients (0.4%) required revisional thoracic outlet decompression. At a median follow-up of 420 days (interquartile range: 150, 937) 93.3% reported symptomatic improvement. CONCLUSION: Based on low composite morbidity, need for very few revisional operations, and high rates of symptomatic improvement, a multidisciplinary treatment approach including primarily supraclavicular thoracic outlet decompression is safe and effective for patients with TOS.

The Increasing Adoption of Minimally Invasive Lobectomy in the United States.

BACKGROUND: The objective of this study is to evaluate the trends of and outcomes associated with the use of minimally invasive lobectomy for stage I and II non-small cell lung cancer (NSCLC) in the United States. METHODS: The use of and outcomes associated with open and minimally invasive lobectomy for clinical stage I and stage II NSCLC from 2010 to 2017 in the National Cancer Database were assessed by multivariable logistic regression and propensity score matching. RESULTS: From 2010 to 2017, use of minimally invasive lobectomies increased for stage I NSCLC (multivariable-adjusted odds ratio [aOR] 4.52; 95% CI, 3.95-5.18; P < .001) and stage II NSCLC (aOR 4.38; 95% CI, 3.38-5.68; P < .001). In 2015, for the first time, more lobectomies for stage I NSCLC were performed by minimally invasive techniques (52.2%, n = 5647) than by thoracotomy (47.8%, n = 5164); and in 2017, more lobectomies for stage II NSCLC were performed by minimally invasive techniques (54.7%, n = 1620) than by thoracotomy (45.3%, n = 1,342). From 2010 to 2017, the conversion rates from minimally invasive to open lobectomy for stage I NSCLC decreased from 19.6% (n = 466) to 7.2% (n = 521; aOR 0.32; 95% CI, 0.23-0.43; P < .001). Similarly, from 2010 to 2017, the conversion rates from minimally invasive to open lobectomy for stage II NSCLC decreased from 20% (n = 114) to 11.5% (n = 186; aOR 0.39; 95% CI, 0.21-0.72; P = .002). CONCLUSIONS: In the United States, for stage I and stage II NSCLC from 2010 to 2017, the use of minimally invasive lobectomy significantly increased while the conversion rate significantly decreased. By 2017, the minimally invasive approach had become the predominant approach for both stage I and stage II NSCLC.

A National Survey of Surgeons Evaluating the Accuracy of Mediastinal Lymph Node Identification.

BACKGROUND: The Commission on Cancer implemented Standard 5.8 in 2021, which requires removal of 3 mediastinal nodes and 1 hilar node with lung cancer resection. We conducted a national survey to assess whether surgeons who treat lung cancer in different clinical settings correctly identify mediastinal lymph node stations. METHODS: Cardiac or thoracic surgeons expressing interest in lung cancer surgery on the Cardiothoracic Surgery Network were asked to complete a 7-question survey assessing their knowledge of lymph node anatomy. General surgeons whose practice includes thoracic surgery were invited through American College of Surgeon's Cancer Research Program. Results were analyzed using Pearson's chi-square test. Multivariable linear regression was used to identify predictors of a higher score on the survey. RESULTS: Of the 280 surgeons that responded, 86.8% were male and 13.2% were female; the median age was 50 years. Of these surgeons, 211 (75.4%) were thoracic, 59 (21.1%) were cardiac, and 10 (3.6%) were general surgeons. Surgeons were most likely to correctly identify lymph node stations 8R and 9R and least likely to correctly identify the midline pretracheal node just superior to the carina (4R). Surgeons whose practice involved a greater percentage of thoracic surgery patients and surgeons who performed a greater number of lobectomies scored higher on the lymph node assessment. CONCLUSION: Knowledge of mediastinal node anatomy among surgeons who perform thoracic surgery is generally high, but varies by clinical setting. Efforts are under way to better educate lung cancer surgeons on nodal anatomy, and to increase adoption of Standard 5.8.

A national analysis of open versus minimally invasive thymectomy for stage I-III thymic carcinoma.

OBJECTIVES: The oncological efficacy of minimally invasive thymectomy for thymic carcinoma is not well characterized. We compared overall survival and short-term outcomes between open and minimally invasive surgical (video-assisted thoracoscopic and robotic) approaches using the National Cancer Database. METHODS: Perioperative outcomes and overall survival of patients who underwent open versus minimally invasive thymectomy for Masaoka stage I-III thymic carcinoma from 2010 to 2015 in the National Cancer Database were evaluated using propensity score-matched analysis and multivariable Cox proportional hazards modelling. Outcomes by surgical approach were assessed using an intent-to-treat analysis. RESULTS: Of the 216 thymectomies that were evaluated, 43 (20%) were performed with minimally invasive techniques (22 video-assisted thoracoscopic and 21 robotic). The minimally invasive approach was associated with a shorter median length of stay when compared to the open approach (3 vs 5 days, P < 0.001). In the propensity score-matched analysis of 30 open and 30 minimally invasive thymectomies, the minimally invasive group did not differ significantly in median length of stay (3 vs 4.5 days, P = 0.27), 30-day readmission (P = 0.13), 30-day mortality (P = 0.60), 90-day mortality (P = 0.60), margin positivity (P = 0.39) and 5-year survival (78.6% vs 54.6%, P = 0.15) when compared to the open group. CONCLUSIONS: In this national analysis, minimally invasive thymectomy for stage I-III thymic carcinoma was found to have no significant differences in short-term outcomes and overall survival when compared to open thymectomy.

A national analysis of open versus minimally invasive thymectomy for stage I to III thymoma.

OBJECTIVE: The oncologic efficacy of minimally invasive thymectomy for thymoma is not well characterized. We compared short-term outcomes and overall survival between open and minimally invasive (video-assisted thoracoscopic and robotic) approaches using the National Cancer Data Base. METHODS: Perioperative outcomes and survival of patients who underwent open versus minimally invasive thymectomy for clinical stage I to III thymoma from 2010 to 2014 in the National Cancer Data Base were evaluated using multivariable Cox proportional hazards modeling and propensity score-matched analysis. Predictors of minimally invasive use were evaluated using multivariable logistic regression. Outcomes of surgical approach were evaluated using an intent-to-treat analysis. RESULTS: Of the 1223 thymectomies that were evaluated, 317 (26%) were performed minimally invasively (141 video-assisted thoracoscopic and 176 robotic). The minimally invasive group had a shorter median length of stay when compared with the open group (3 [2-4] days vs 4 [3-6] days, P < .001). In a propensity score-matched analysis of 185 open and 185 minimally invasive (video-assisted thoracoscopic + robotic) thymectomy, the minimally invasive group continued to have a shorter median length of stay (3 vs 4 days, P < .01) but did not have significant differences in margin positivity (P = .84), 30-day readmission (P = .28), 30-day mortality (P = .60), and 5-year survival (89.4% vs 81.6%, P = .20) when compared with the open group. CONCLUSIONS: In this national analysis, minimally invasive thymectomy was associated with shorter length of stay and was not associated with increased margin positivity, perioperative mortality, 30-day readmission rate, or reduced overall survival when compared with open thymectomy.