Prediction of Postoperative Pulmonary Complications in Thoracic Surgery
For patients and families
In plain language
An automatic summary of structured registry data. It is an orientation aid, not a substitute for the official protocol or a physician assessment.
- What is being studied
- The protocol lists: Evaluation of prognostic performance of a defined score using a machine learning method (STABL: Stability Selection) integrating immune data (cytometric and proteomic).
- Who it may be relevant to
- Registry conditions: Postoperative Pulmonary Complications (PPCs). Basic parameters: 18 years — 99 years · All.
- What needs checking
- Age, condition and sex are only basic indicators. Prior treatment, laboratory values and other mandatory requirements appear in the eligibility criteria below.
- Where it takes place
- France
- Next step
- Save the trial, show it to the treating physician, and confirm current recruitment with the study center. Costs, documents and travel →
Unsure about the terms? Read our patient guide →
Official title
Prediction of Postoperative Pulmonary Complications in Thoracic Surgery: an Immuno-inflammatory Approach
Overview
Lung cancer is a common disease, and its treatment is lobectomy or pulmonary segmentectomy. In France, approximately 8,000 patients undergo this procedure each year, but it remains associated with significant Postoperative Pulmonary Complications (PPC). This surgical trauma triggers a multicellular and orchestrated immune response, necessary for defense against pathogens, as well as for inflammatory resolution and wound healing. Preoperative single-cell analysis of the patient's immune system is therefore a promising strategy for identifying biomarkers of postoperative pulmonary complications (PPC). Brice Gaudilliere's laboratory at Stanford University, in collaboration with the Paris-based startup Surge, has developed and patented a multivariate model integrating mass cytometry data, proteomic analyses, and clinical data collected before surgery to accurately predict surgical site complications after major abdominal surgery. However, no study has yet explored the identification of inflammatory biomarkers predictive of PPC after thoracic surgery.
Detailed description
The issue of postoperative pulmonary complications following major lung resection (such as lobectomy or segmentectomy) is a central topic in anesthesia and thoracic surgery. Postoperative morbidity and mortality after this type of surgery have drastically decreased in recent years with advances in anesthesia and resuscitation, as well as minimally invasive surgery, but remain high compared to other types of surgery, particularly due to postoperative pneumonia. The etiology of postoperative pneumonia is multifactorial (atelectasis, postoperative ventilation, inadequate analgesia), but the patient's immune system plays a predominant role in each individual case. Therefore, identifying inflammatory biomarkers predictive of postoperative pulmonary complications in a given patient could optimize their management and reduce the risk of postoperative pulmonary cancer (PPC). The objective of this study is to identify preoperative inflammatory biomarkers predictive of PPC after major lung resection. It will use machine learning methods specific to these data to define an immune signature of PPC. This immune signature will be validated using standard analytical techniques to facilitate the clinical translation of a diagnostic test.
Interventions
- Diagnostic test Evaluation of prognostic performance of a defined score using a machine learning method (STABL: Stability Selection) integrating immune data (cytometric and proteomic)
Determination of the area under the curve (AUC) Receiver Operating Curve (ROC) for predicting complications calculated from the score obtained by the machine learning method and the occurrence of at least one major pulmonary complication among the following in the first 7 postoperative days: postoperative pneumonia, pleural effusion, postoperative atelectasis, pneumothorax, bronchospasm and acute respiratory distress syndrome.
Primary outcome measures
- Evaluation of the prognostic performance of a score for screening patients at risk of postoperative pulmonary complications (PPC) [Time frame: Evaluation of the prognostic performance of a defined score using a machine learning method (STABL: Stability Selection) integrating preoperative immune (cytometric and proteomic) and clinical data within 7 postoperative days of a major lung resection]
Secondary outcome measures (12)
- Evaluation of the incidence of pulmonary complications [Time frame: 30 days]
- Evaluation of the correlation between the prognostic score defined using a machine learning method and the length of hospital stay [Time frame: 3 months]
- Evaluation of the correlation between the prognostic score defined using a machine learning method and the number of reintubations recorded [Time frame: 30 days]
- Evaluation of the correlation between the prognostic score defined using a machine learning method and the Number of unplanned hospitalizations in intensive care recorded [Time frame: 30 days]
- Evaluation of the correlation between the prognostic score defined using a machine learning method and the Preoperative anxiety score assessed [Time frame: 48 hours]
- Evaluation of the correlation between the prognostic score defined using a machine learning method and the Preoperative anxiety score assessed [Time frame: 48 hours]
- Evaluation of the correlation between the prognostic score defined using a machine learning method and The cost of care [Time frame: 3 months]
- Evaluation of the prognostic performance of the score calculated by the machine learning method on Post-operative Pulmonary Complications (PPC) assessed by the Melbourne composite score (Melbourne Group Scale (MGS) >=4) [Time frame: 7 days]
- Evaluation of the prognostic performance of the score calculated by the machine learning method on the severity of postpartum bleeding (PPB) [Time frame: 30 days]
- Evaluation of the prognostic performance of the score calculated by the machine learning method on Postoperative mortality assessed at 30 days [Time frame: 30 days]
- Evaluation of the prognostic performance of the score calculated by the machine learning method on Postoperative mortality assessed at 90 days [Time frame: 90 days]
- Evaluation of the prognostic performance of the score calculated by the machine learning method on Pre- and postoperative pain [Time frame: 90 days]
Eligibility criteria
Inclusion criteria
- Age ≥ 18 years
- ASA score ≤ 3
- Patients undergoing scheduled video-assisted or robot-assisted lobectomy, bilobectomy, or segmentectomy.
- Patients who have read and understood the information letter and do not object to the research.
- For women of childbearing age (non-sterile): effective contraception
- Menopausal (non-medically induced amenorrhea for at least 12 months)
- Patients covered by a social security scheme
Exclusion criteria
- Minor patients
- Surgery scheduled for a Friday
- Patients undergoing a pneumonectomy
- Pregnant or breastfeeding women
- Patients deprived of their liberty by an administrative or judicial decision, as well as those under legal protection, guardianship, or curatorship
Criteria are shown verbatim from the registry (in English). Final eligibility is always assessed by the study center.
Healthy volunteers: No
Study design
- Observational model
- Case-only
Study locations
France · 1 center
- Service de Anesthésie-Réanimation Médecine périopératoire CHU de Rouen — Rouen
Identifiers
NCT: NCT07359885 · 2024/0307/HP · 2025-A01699-40