Improving pleural procedures: training, image analysis and pleural manometry

Salamonsen, Matthew (2015). Improving pleural procedures: training, image analysis and pleural manometry PhD Thesis, School of Medicine, The University of Queensland. doi:10.14264/uql.2015.695

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s4261116_phd_submission.pdf Thesis (open access) application/pdf 6.61MB 151

Author Salamonsen, Matthew
Thesis Title Improving pleural procedures: training, image analysis and pleural manometry
School, Centre or Institute School of Medicine
Institution The University of Queensland
DOI 10.14264/uql.2015.695
Publication date 2015-06-05
Thesis type PhD Thesis
Open Access Status Other
Supervisor David Fielding
Karin Steinke
Total pages 122
Language eng
Subjects 110399 Clinical Sciences not elsewhere classified
110203 Respiratory Diseases
119999 Medical and Health Sciences not elsewhere classified
Formatted abstract
Background Adverse events related to thoracentesis and chest tube insertion are common. International guidelines recommend improved procedural training and mandatory use of ultrasound (US). A particular problem can arise when managing a malignant pleural effusion. If lung entrapment is present, this changes the optimal treatment strategy.

Aims To increase the safety and therapeutic efficacy of pleural procedures, by improving training and extending the utility of US and pleural manometry to guide these procedures.

Methods The aims will be achieved through three study arms.
1. Assessment tools: Objective tools to assess the competency of physician-performed chest tube insertion (the Chest Tube Insertion Competency Test: TUBE-iCOMPT) and thoracic US (the Ultrasound-Guided Skills and Tasks Assessment Test: UG-STAT) were written in line with international evidence-based guidelines. The reliability and accuracy of the tools was demonstrated by applying them to assess a cohort of doctors performing thoracic US and chest tube insertion.

2. US of the intercostal artery (ICA): A method to visualise the ICA with US was initially developed on a cohort of patients undergoing routine thoracic ultrasound prior to thoracentesis. The sensitivity and specificity of US to screen for a vulnerable intercostal artery was then assessed in a second study, by comparing the position of the ICA found by US with that on computed tomography (CT) chest examination.

3i. Pleural manometry: A new method for continuous pleural manometry during thoracentesis (compared with the traditional intermittent method) was developed. By transducing pleural pressures from an epidural catheter sitting within the pleural effusion, pressure measurements can be made without halting the flow of fluid through the drainage catheter. On a cohort of patients undergoing thoracentesis, pleural manometry was performed simultaneously using the traditional and the new method. Differences in opening pressure and pleural elastance derived from each technique were compared, before demonstrating the feasibility of fully automated pleural manometry allowed by the new technique, by connecting the system to a urodynamics machine.

3ii. US and entrapped lung: A method to identify malignant entrapped lung with US prior to pleural effusion drainage was developed and assessed. Thoracic US with an echocardiogram machine was performed on a cohort of patients prior to thoracentesis. The displacement (mm) and deformation (%) of the atelectatic lower lobe due to the transmitted cardiac impulse, was measured using motion mode (M Mode) and speckle-tracking imaging (STI) respectively. The gold standard diagnosis of lower lobe entrapment was made according to post-drainage radiology. Data was randomly divided into a development and validation set. Data from the development set was used to construct receiver-operating curves (ROC), which described the ability of US to identify lung entrapment. Optimal diagnostic cut-offs selected from the ROC curves were then applied to the validation set, and diagnostic indices calculated.

1. Both the TUBE-iCOMPT and UG-STAT showed a high degree of test-retest and inter-tester reliability, and were able to appropriately separate participants into beginner, intermediate and advanced groups.

2. Ultrasound was able to identify a vulnerable intercostal artery with a sensitivity of 86% and specificity of 30%. The performance of a high-end machine was not significantly better than a mobile machine, and an appropriately trained respiratory physician was as good as an experienced radiographer.

3i. There was no statistical difference in opening pressures (p=0.49) or pleural elastance (p>0.10) derived from each technique. The new technique allowed for fully automated real-time display of pleural pressures, drainage volumes and pleural elastance.

3ii. Motion analysis with US showed good diagnostic power to identify malignant entrapped lung prior to pleural effusion drainage, with an area under the ROC of 0.79 for M Mode and 0.86 for STI. The sensitivity and specificity were 50% and 85% for M Mode and 71% and 85% for STI respectively.

This PhD thesis has produced studies that will have a real impact on clinical practice. It has developed objective assessment tools with initial pilot studies demonstrating their validity. They could now provide the impetus for further studies to develop and refine such methods for procedural competency assessment. While these studies are underway individual hospitals could use the tools to track an individual’s progress through training and use these assessments as feedback opportunities. It has also produced novel extensions to the current use of ultrasound by the pulmonologist, documenting for the first time the possible use of US to identify a vulnerable ICA prior to chest drainage procedures and prospectively identify malignant lung entrapment. It has refined the current technique of pleural manometry, simplifying the practice for clinicians and allowing for fully automated recording and calculation of relevant parameters. These novel tools now provide the stimulus to further clinical study as outlined above. In time it is hoped these tools may improve outcomes and reduce adverse events suffered by our patients.
Keyword Pleural effusion
Intercostal catheter
Procedural training
Pleural malignancy
Pleural manometry
Lung entrapment
Trapped lung

Document type: Thesis
Collections: UQ Theses (RHD) - Official
UQ Theses (RHD) - Open Access
Version Filter Type
Citation counts: Google Scholar Search Google Scholar
Created: Sun, 24 May 2015, 14:43:35 EST by Matthew Salamonsen on behalf of School of Medicine