Predicting factors and impact of abnormal preoperative chest radiograph inelective non-cardiothoracic surgery: a prospective observational study
5
5
Issued Date
2020
Copyright Date
2020
Resource Type
Language
eng
File Type
application/pdf
No. of Pages/File Size
xxii, 165 leaves ; 30 cm.
Access Rights
open access
Rights
ผลงานนี้เป็นลิขสิทธิ์ของมหาวิทยาลัยมหิดล ขอสงวนไว้สำหรับเพื่อการศึกษาเท่านั้น ต้องอ้างอิงแหล่งที่มา ห้ามดัดแปลงเนื้อหา และห้ามนำไปใช้เพื่อการค้า
Rights Holder(s)
Mahidol University
Bibliographic Citation
Thesis (Ph.D. (Clinical Epidemiology))--Mahidol University, 2020
Suggested Citation
Alisa Seangleulur Predicting factors and impact of abnormal preoperative chest radiograph inelective non-cardiothoracic surgery: a prospective observational study. Thesis (Ph.D. (Clinical Epidemiology))--Mahidol University, 2020. Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/117632
Title
Predicting factors and impact of abnormal preoperative chest radiograph inelective non-cardiothoracic surgery: a prospective observational study
Author(s)
Abstract
Background: Although many guidelines have recommended to order preoperative chest radiograph (POCR), it still varied practice among physicians and the validity of guidelines in tuberculosis-burden country is controversial. Ordering POCRs for all patients who undergo to operation may not be useful but may be beneficial to some risk patients. This study aimed to determine the prevalence of significant abnormal POCR, identify predictors, and develop the prediction score to classify a risk of significant abnormal POCR. Methods: A cross-sectional study was conducted which consisted of derivative phase at Ramathibodi hospital (RAMA) and validation phase at Thammasat University hospital (TUH) in Thailand during February and March 2020. Patients underwent elective non-cardiothoracic surgery and required anesthesia were included for the study. Multiple logistic regression was applied to develop a parsimonious model. Performances of the model were internally and externally assessed. The consequences of abnormal POCRs in perioperative management and postoperative complications were also assessed at TUH. Study had been approved by the ethical committee of both hospitals. Results: A total of 703 and 403 patients at RAMA and TUH were respectively included with the prevalence of abnormal POCRs of 19.5% (16.6%, 22.6%) and 21.2% (17.3%, 25.4%). Four predictors (i.e. age, preexisting pulmonary disease, preexisting cardiac disease, and diabetes mellitus) were significantly associated with significant abnormal POCR, and thus included in the parsimonious model. The model showed good performance in calibration and discrimination from both external and internal validations. For internal validation, the mean (95% CI) bootstrap corrected calibration and discrimination coefficient was 0.465 (0.463, 0.466) and 0.732 (0.732, 0.733), respectively. For external validation, the O/E ratio was 0.987 (0.888, 1.085), and the C-statistics was 0.731 (0.674, 0.789). Risk prediction scores were classified into 4 risk groups indicating intermediate to high risk patients would be about 32.3% (27.1%, 38.0%) having significant abnormal POCRs. Twenty-nine patients [7.1% (4.8%, 10.0%)] had abnormal POCR which affected on their perioperative managements. Conclusion: The risk prediction model of significant abnormal POCR was developed with good calibration and discrimination. The model might be useful for selecting patients to receive POCR before surgery.
Degree Name
Doctor of Philosophy
Degree Level
Doctoral degree
Degree Department
Faculty of Science
Degree Discipline
Clinical Epidemiology
Degree Grantor(s)
Mahidol University