Borja Aguinagalde

Objective: To perform an external validation of a previously reported machine learning (ML) approach for predicting the diagnosis of pleural tuberculosis. Patients and Methods: We defined two cohorts: a Training group, comprising 273 out of 1,220 effusions from our prospective study (2013–2022); and a Testing group, from a retrospective analysis of 360 effusions from 832 consecutive patients in Bajo Deba health district (1996–2012). All the effusions included were exudative and lymphocytic. In Training and Testing groups respectively, 49 and 104 cases were tuberculous, 143 and 92 were malignant, and 81 and 164 were diagnosed with “other diseases”; pre-test probabilities of pleural tuberculosis were 4% and 12.7%. Variables included were: age, pH, adenosine deaminase, glucose, protein, and lactate dehydrogenase levels, and white cell counts (total and differential) in pleural fluid. We used two ML classifiers: binary (tuberculous and non-tuberculous), and three-class (tuberculous, malignant, and others); and compared them with Bayesian analysis. Results: The best binary classifier yielded a sensitivity of 88%, specificity of 98%, and accuracy of 95%. The best three-class classifier achieved the same accuracy and correctly classified 83% (77/92) of malignant cases. The ML models yielded higher positive predictive values than Bayesian analysis based on ADA > 40 U/l and lymphocyte percentage ≥ 50% (92%). Conclusions: This external validation confirms the good performance of the previously reported ML approach for predicting the diagnosis of pleural tuberculosis based on exudative and lymphocytic pleural effusions, and for discriminating the cases most likely to be malignant. Additionally, ML was more accurate than the Bayesian approach in our study.

Background: Surgical resection remains the standard treatment for early-stage non-small-cell lung cancer (NSCLC). Traditionally, lobectomy has been considered the procedure of choice; however, emerging evidence suggests that trisegmentectomy may offer comparable outcomes. This meta-analysis evaluates whether left upper lobe trisegmentectomy provides non-inferior or superior oncologic outcomes compared to left upper lobectomy, with particular attention to recurrence patterns. Methods: Following PRISMA guidelines, we included comparative studies evaluating left upper lobectomy versus trisegmentectomy. Outcomes assessed included recurrence (locoregional and distant), morbidity, and the length of hospital stay. A meta-analysis was conducted using the metabin function from the R meta package. Results: Of 14 identified articles, 9 met the inclusion criteria. No significant differences were observed in locoregional recurrence. However, distant recurrence was significantly lower in the trisegmentectomy group (OR 0.58; 95% CI 0.41–0.82). While overall morbidity showed no significant difference (OR 0.95), analysis of matched studies favored trisegmentectomy (OR 0.73; 95% CI 0.56–0.96). Hospital stay was significantly shorter in the trisegmentectomy group (OR –0.94; 95% CI –1.26 to –0.63). Conclusions: Trisegmentectomy and lobectomy exhibit distinct recurrence patterns, with lobectomy associated with a higher rate of distant recurrence. Trisegmentectomy may provide oncologic and perioperative advantages in appropriately selected patients. The systematic review and meta-analysis are registered in PROSPERO (registration number: CRD420251066445).

Background/Objectives: The objective was to compare the results of segmentectomy and lobectomy in the treatment of c-stage IA lung cancer in terms of tumor recurrence and 5-year survival. Methods: An observational study was performed using 3533 patients included in the registry of the Spanish VATS Group (GEVATS) of the Spanish Society of Thoracic Surgery (SECT) between 2016 and 2018. A total of 1004 lobectomies and 83 segmentectomies in c-stage IA were selected. Two comparable groups were selected through 2:1 propensity score matching with patient-, tumor- and surgery-related variables, leaving 166 lobectomies and 83 segmentectomies. Tumor recurrence was analyzed by Fisher’s test and overall, cancer-specific, recurrence-free and disease-free survival by Kaplan-Meier and Log-rank tests. Results: Overall recurrence was 23.7% in both groups, with a predominance of locoregional recurrence in segmentectomy (16.2% vs. 11.2%) and distant recurrence in lobectomy (12.5% vs. 7.5%). There was no difference between the two groups in any of the survival types. Overall survival at 5 years was 73.5% (95% CI: 65.5–82.4%) in the lobectomy group vs. 73.1% (95% CI: 60.1–88.9%) in the segmentectomy group. Conclusions: Anatomic segmentectomy may be a valid option in the treatment of c-stage IA lung cancer since the recurrence and long-term survival outcomes compared to lobectomy are equivalent.

Electronic patient-reported outcome-based symptom management, a new era in surgical patient management

Background/Objectives: The aim of this study was to analyze whether the implementation of artificial intelligence (AI), specifically the Natural Language Processing (NLP) branch developed by OpenAI, could help a thoracic multidisciplinary tumor board (MTB) make decisions if provided with all of the patient data presented to the committee and supported by accepted clinical practice guidelines. Methods: This is a retrospective comparative study. The inclusion criteria were defined as all patients who presented at the thoracic MTB with a suspicious or first diagnosis of non-small-cell lung cancer between January 2023 and June 2023. Intervention: GPT 3.5 turbo chat was used, providing the clinical case summary presented in committee proceedings and the latest SEPAR lung cancer treatment guidelines. The application was asked to issue one of the following recommendations: follow-up, surgery, chemotherapy, radiotherapy, or chemoradiotherapy. Statistical analysis: A concordance analysis was performed by measuring the Kappa coefficient to evaluate the consistency between the results of the AI and the committee’s decision. Results: Fifty-two patients were included in the study. The AI had an overall concordance of 76%, with a Kappa index of 0.59 and a consistency and replicability of 92.3% for the patients in whom it recommended surgery (after repeating the cases four times). Conclusions: AI is an interesting tool which could help in decision making in MTBs.

Background: Mediastinal lymph node staging is a key element in the diagnosis of lung cancer. The combination of computed tomography (CT) and positron emission tomography (PET) has improved staging but some circumstances are known to influence their negative predictive value. The objective of this study was to assess the impact on survival of avoiding invasive mediastinal staging in surgical lung cancer patients with negative mediastinum in CT and PET and intermediate risk of unexpected pN2. Methods: Data were collected from the prospective cohort of the Spanish Group for Video-Assisted Thoracic Surgery (GEVATS), from December 2016 to March 2018. For this study, patients were selected if they had negative mediastinum in CT and PET findings but tumours >3 cm or located centrally, or with cN1 disease. Patients who did and did not undergo invasive staging [invasive group (IG) and non-invasive group (NIG)] were compared, analysing unexpected pN2 and survival with Kaplan-Meier curves and Cox regression. Results: A total of 2,826 patients underwent surgery for primary lung cancer. We selected 1,247 patients who had tumours >3 cm, central tumours or cN1. Invasive staging was performed in 275 (22.1%) cases. The unexpected pN2 rate was 9.6% in the NIG and 13.8% in the IG, but half of them were discovered prior to surgery in the IG. Five-year overall survival (OS) was poorer in the IG (52.4% vs. 64%; PConclusions: Invasive staging recommended by guidelines could be reduced with an appropriate selection in mediastinal CT- and PET-negative patients with risk factors for unexpected pN2, because rates of pN2 and survival did not worsen without invasive staging.

There is no consensus in the literature on preoperative histological analysis for lung cancer. The objective of this study was to assess 4 diagnostic models used in different hospitals with differing practices regarding preoperative histological diagnosis and the consequences in terms of unnecessary surgery and futile major resection. We carried out a retrospective observational study collected from 4 university hospitals in Spain over 3 years (January 2019 to December 2021). We included all patients with a confirmed diagnosis of primary lung cancer and any patients with suspected primary lung cancer who had undergone surgery. All patients underwent computed tomography and positron emission tomography/computed tomography scans. Each multidisciplinary committee was free to choose whether to perform flexible bronchoscopic or transthoracic lung biopsy. Decisions concerning whether to perform intraoperative sample analysis, the surgical approach and the type of resection were left to the surgical team. We included a total of 1642 patients. The use of flexible endoscopy and its diagnostic performance varied substantially between hospitals (range: 23.8–79.3% and 25–60.7%, respectively); and the same was observed for transthoracic biopsy and its performance (range: 16.9–82.3% and 64.6–97%, respectively). Regarding major resection surgery (lobectomy or more extensive resection), the lowest rate was observed in hospital C (1%) and the highest in hospital B (2.8%), with between-hospital differences not reaching significance (P = 0.173). The rate of histological sampling before lung cancer surgery still varies between hospitals. In spite of very diverse multidisciplinary management, the rate of futile lobectomy is not significantly higher in hospitals with lower rates of preoperative histological analysis.

Background: Lung resection using video-assisted thoracoscopic surgery (VATS) improves surgical accuracy and postoperative recovery. Unfortunately, moderate-to-severe acute postoperative pain is still inherent to the procedure, and a technique of choice has not been established for the appropriate control of pain. In this study, we aimed to compare the efficacy and safety of intrathecal morphine (ITM) with that of intercostal levobupivacaine (ICL). Methods: We conducted a single-center, prospective, randomized, observer-blinded, controlled trial among 181 adult patients undergoing VATS (ISRCTN12771155). Participants were randomized to receive ITM or ICL. Primary outcomes were the intensity of pain, assessed by a numeric rating scale (NRS) over the first 48 h after surgery, and the amount of intravenous morphine used. Secondary outcomes included the incidence of adverse effects, length of hospital stay, mortality, and chronic post-surgical pain at 6 and 12 months after surgery. Results: There are no statistically significant differences between ITM and ICL groups in pain intensity and evolution at rest. In cough-related pain, differences in pain trajectories over time are observed. Upon admission to the PACU, cough-related pain was higher in the ITM group, but the trend reversed after 6 h. There are no significant differences in adverse effects. The rate of chronic pain was low and did not differ significantly between groups. Conclusions: ITM can be considered an adequate and satisfactory regional technique for the control of acute postoperative pain in VATS, compatible with the multimodal rehabilitation and early discharge protocols used in these types of surgeries.