Outcome Analysis of Thoracic Surgeries: A Retrospective Study Using Extensive Patient Data

Abstract

Introduction: Thoracic procedures include substantial risks and complications. Examining postoperative outcomes is essential for comprehending the efficacy of treatment and pinpointing potential risk factors. This study examined a comprehensive dataset of patients who underwent thoracic surgery in order to evaluate the outcomes and their correlations.
Methods: Retrospective study was performed on a dataset consisting of 470 patients who undergone different types of thoracic operations. An analysis was conducted on variables related to the patient's characteristics before surgery (such as demographics and comorbidities), the type of procedure performed during surgery, and any complications or death that occurred after surgery. The sample was characterized using descriptive statistics. The relationships between variables were evaluated through the use of correlation analysis, analysis of variance (ANOVA), chi-square test, and Bayesian regression analysis.
Results: The study consisted of 349 lobectomies, 47 pneumonectomies, and surgeries for diverse illnesses. The average age of the patients was 62.53 years. The rate of death within one year after surgery was 14.9%. Greater procedural risk scores were linked to higher fatality rates (Bayesian regression analysis revealed a posterior mode of 3.682, with a 95% confidence interval of 0.618-6.746). Lower forced expiratory volume was found to be a predictive factor for worse outcomes in the preoperative stage. The mortality rate for procedure type DGN2/3 was lower than that for DGN4/5/6, with posterior modes ranging from 0.690 to 3.639 and 95% confidence intervals that overlapped with zero.
Conclusion: The examination of a substantial dataset on thoracic surgery yielded valuable information about the results and their correlations. Elevated risk ratings and certain preoperative variables were associated with higher fatality rates. Lobectomies yielded superior outcomes compared to large resections. The utilization of comprehensive patient-level data and the application of a Bayesian methodology facilitate the comprehension of the elements that influence the effectiveness of thoracic surgery. Prospective studies have the ability to investigate high-risk subgroups that have been identified.