Plain and computed radiography for detecting experimentally induced pneumothorax in cadavers: implications for detection in patients

Deep Learning for Detection of Pneumothorax and Pleural Effusion on Chest Radiographs: Validation Against Computed Tomography, Impact on Resident Reading Time, and Interreader Concordance

PURPOSE

To study the performance of artificial intelligence (AI) for detecting pleural pathology on chest radiographs (CXRs) using computed tomography as ground truth.

PATIENTS AND METHODS

Retrospective study of subjects undergoing CXR in various clinical settings. Computed tomography obtained within 24 hours of the CXR was used to volumetrically quantify pleural effusions (PEfs) and pneumothoraxes (Ptxs). CXR was evaluated by AI software (INSIGHT CXR; Lunit) and by 3 second-year radiology residents, followed by AI-assisted reassessment after a 3-month washout period. We used the area under the receiver operating characteristics curve (AUROC) to assess AI versus residents' performance and mixed-model analyses to investigate differences in reading time and interreader concordance.

RESULTS

There were 96 control subjects, 165 with PEf, and 101 with Ptx. AI-AUROC was noninferior to aggregate resident-AUROC for PEf (0.82 vs 0.86, P < 0.001) and Ptx (0.80 vs 0.84, P = 0.001) detection. AI-assisted resident-AUROC was higher but not significantly different from the baseline. AI-assisted reading time was reduced by 49% (157 vs 80 s per case, P = 0.009), and Fleiss kappa for Ptx detection increased from 0.70 to 0.78 ( P = 0.003). AI decreased detection error for PEf (odds ratio = 0.74, P = 0.024) and Ptx (odds ratio = 0.39, P < 0.001).

CONCLUSION

Current AI technology for the detection of PEf and Ptx on CXR was noninferior to second-year resident performance and could help decrease reading time and detection error.

Effect of Contrast Level and Image Format on a Deep Learning Algorithm for the Detection of Pneumothorax with Chest Radiography

Under the black-box nature in the deep learning model, it is uncertain how the change in contrast level and format affects the performance. We aimed to investigate the effect of contrast level and image format on the effectiveness of deep learning for diagnosing pneumothorax on chest radiographs. We collected 3316 images (1016 pneumothorax and 2300 normal images), and all images were set to the standard contrast level (100%) and stored in the Digital Imaging and Communication in Medicine and Joint Photographic Experts Group (JPEG) formats. Data were randomly separated into 80% of training and 20% of test sets, and the contrast of images in the test set was changed to 5 levels (50%, 75%, 100%, 125%, and 150%). We trained the model to detect pneumothorax using ResNet-50 with 100% level images and tested with 5-level images in the two formats. While comparing the overall performance between each contrast level in the two formats, the area under the receiver-operating characteristic curve (AUC) was significantly different (all p < 0.001) except between 125 and 150% in JPEG format (p = 0.382). When comparing the two formats at same contrast levels, AUC was significantly different (all p < 0.001) except 50% and 100% (p = 0.079 and p = 0.082, respectively). The contrast level and format of medical images could influence the performance of the deep learning model. It is required to train with various contrast levels and formats of image, and further image processing for improvement and maintenance of the performance.

Bedside electrical impedance tomography in early diagnosis of pneumothorax in mechanically ventilated ICU patients — a single-center retrospective cohort study

PURPOSE

This study aimed to evaluate the routine use of electrical impedance tomography (EIT) to diagnose pneumothorax (PTX) in mechanically ventilated patients in the intensive care unit (ICU).

METHODS

A retrospective cohort study was conducted including mechanically ventilated supine patients who received EIT examinations. The EIT-based tidal variation was divided into ventral and dorsal regions of interest (ROIs): upper right (UR, ROI1), upper left (UL, ROI2) lower right (LR, ROI3), and lower left (LL, ROI4), and the ventilation defect score (DS) was calculated in each quadrant. Furthermore, horizontal ventral ventilation index (HVVI) was defined as ROI1_% / ROI2_% in the two ventral quadrants if ROI1_% > ROI2_%, otherwise HVVI = ROI2_% / ROI1_%.

RESULTS

A total of 203 patients were included, 25 of them with confirmed PTX. In the PTX patients, preceding cardiac surgery was the most common cause of PTX. Compared with the patients without PTX, the PTX patients had a higher DS in the ventral quadrants [median and interquartile range (IQR): 1.00 (0.00, 2.00) vs. 0.00 (0.00, 0.00), P < 0.001] respectively, but similar in the dorsal quadrants [median and IQR: 1.00 (0.00, 1.00) vs. 0.00 (0.00, 1.00), P = 0.722]. Moreover, a higher HVVI was found in the PTX group [median and IQR: 2.51 (1.58, 3.52) vs. 1.36 (1.15, 1.77), P < 0.001]. The area under the receiver operating characteristic curve of the HVVI to differentiate PTX from non-PTX was 0.88, with a sensitivity of 70% and a specificity of 90% when the cut-off value was 2.57.

CONCLUSION

The ventilation defect in the ventral regions and a high HVVI on EIT were observed in mechanically ventilated patients with PTX, which should trigger further diagnostics to confirm it.

Early air transport after thoracic surgery might be safe: A retrospective observational study in the French Caribbean

Objective

The objective of this study was to determine the incidence of early air transport (EAT) morbidity after transpleural surgery. We compared our cohort with our patients not requiring air transport.

Methods

This was a retrospective observational study, in the Thoracic and Cardiovascular Surgery Department of the University Hospital of Martinique over 40 months. We included all of the files (national and local database, and systematic postoperative consultation) of patients operated on for thoracic surgery or distinguished transpleural surgical intervention, whatever their geographical origin. Patients from another French department benefited from EAT. The complications were classified according to Clavien–Dindo before or after the EAT. Diagnostic criteria were chest pain, dyspnea, and abnormal chest radiograph. Continuous variables are presented as mean, median, and SDs. Discrete variables are presented as n (%).

Results

Of 491 patients operated on, 315 were transpleural surgeries, and 99 patients benefited from EAT. There were 55% resections, a percent predicted of forced expiratory volume in 1 second, and an average preoperative Tiffeneau ratio of respectively, 86% and 78. One complication was found: a pneumothorax in an emphysematous patient, 15 days after the flight, who had an index of prolonged air leak >10. The mean time between surgery and flight was 7.2 days (σ = 4.5), and 3.3 days (σ = 2.9) between removal of the last drain and flight. The morbidity of EAT after transpleural surgery was 1%. The 2 cohorts of “EAT” and “Locals” patients were statistically comparable, particularly in morbidity.

Conclusions

EAT appears to be safe after transpleural surgery, following usual criteria for hospital discharge. It would be interesting to study, on a larger scale, the effect of IPAL as an independent risk factor (in case of high IPAL > 10) as well as pathologies that modify transpleural pressures restrictive ventilatory defect.

Deep multi-instance transfer learning for pneumothorax classification in chest X-ray images

PURPOSE

Pneumothorax is a life-threatening emergency that requires immediate treatment. Frontal-view chest X-ray images are typically used for pneumothorax detection in clinical practice. However, manual review of radiographs is time-consuming, labor-intensive, and highly dependent on the experience of radiologists, which may lead to misdiagnosis. Here, we aim to develop a reliable automatic classification method to assist radiologists in rapidly and accurately diagnosing pneumothorax in frontal chest radiographs.

METHODS

A novel residual neural network (ResNet)-based two-stage deep-learning strategy is proposed for pneumothorax identification: local feature learning (LFL) followed by global multi-instance learning (GMIL). Most of the nonlesion regions in the images are removed for learning discriminative features. Two datasets are used for large-scale validation: a private dataset (27 955 frontal-view chest X-ray images) and a public dataset (the National Institutes of Health [NIH] ChestX-ray14; 112 120 frontal-view X-ray images). The model performance of the identification was evaluated using the accuracy, precision, recall, specificity, F1-score, receiver operating characteristic (ROC), and area under ROC curve (AUC). Fivefold cross-validation is conducted on the datasets, and then the mean and standard deviation of the above-mentioned metrics are calculated to assess the overall performance of the model.

RESULTS

The experimental results demonstrate that the proposed learning strategy can achieve state-of-the-art performance on the NIH dataset with an accuracy, AUC, precision, recall, specificity, and F1-score of 94.4% ± 0.7%, 97.3% ± 0.5%, 94.2% ± 0.3%, 94.6% ± 1.5%, 94.2% ± 0.4%, and 94.4% ± 0.7%, respectively.

CONCLUSIONS

The experimental results demonstrate that our proposed CAD system is an efficient assistive tool in the identification of pneumothorax.

A Retrospective Analysis of the Effectiveness of Extrapleural Autologous Blood Patch Injection on Pneumothorax and Intervention Need in CT-guided Lung Biopsy

PURPOSE

To assess the effect of extrapleural autologous blood injection (EPABI) technique on pneumothorax development before and after coaxial needle withdrawal (CNW) and intervention rate for pneumothorax. To analyze the risk factors of pneumothorax and parenchymal hemorrhage.

MATERIALS AND METHODS

The records of 288 patients who had lung biopsies were analyzed. Of these patients, 188 received EPABI (group-A) before penetrating the parietal pleura, and the remaining did not (group-B). Intraparenchymal autologous blood patch injection was applied at the end of the procedure. The pneumothorax rates before/after CNW and intervention requirement for pneumothorax were compared between groups. The risk factors of pneumothorax before/after CNW and parenchymal hemorrhage were assessed with stepwise logistic regression.

RESULTS

The pneumothorax rate before CNW was significantly lower in group-A (5.92%) than in group-B (19.10%) (p = 0.029). Pneumothorax risk before CNW was reduced if EPABI was applied and skin-to-pleura distance increased. The pneumothorax rate after CNW was similar between two groups (group-A: 6.94%, group-B: 8%), while emphysema grade along the needle path and procedure duration was the significant risk factor. The intervention requirement for pneumothorax was significantly lower in group-A (6.38%) than in group-B (16%) (p = 0.012). Needle aspiration requirement was significantly reduced in group-A. The rate of external drainage catheter and chest tube placement was similar in both groups. The risk factors of parenchymal hemorrhage were overall emphysema grade of the lung, target-to-pleura distance, and target size.

CONCLUSION

Use of EPABI along with IAPBI significantly decreased the pneumothorax rate during biopsy procedure and the intervention rate compared to IAPBI-alone.

Novel Use of Extrapleural Autologous Blood Injection in CT-Guided Percutaneous Lung Biopsy and its Comparison to Intraparenchymal Autologous Blood Patch Injection: A Single-Center, Prospective, Randomized, and Controlled Clinical Trial

PURPOSE

To evaluate the rate of iatrogenic pneumothorax and the need for intervention with extrapleural autologous blood injection (EPABI) along with intraparenchymal autologous blood patch injection (IABPI) or IABPI-only in CT-guided percutaneous lung biopsy.

MATERIALS AND METHODS

One hundred and thirty-nine participants were referred for CT-guided percutaneous lung biopsy, and 81 were randomized into study (EPABI + IABPI, n = 40) and control (IABPI-only, n = 41) groups. In the study group, ~5 ml of autologous blood was injected into the extrapleural space before passing through the parietal and visceral pleura. The primary outcome was the incidence of pneumothorax in two cohorts within 2 weeks after the procedure.

RESULTS

In the per-protocol population, pneumothorax rates were 5.9% and 25.7% in the study and control groups, respectively. The difference between the two groups was -19.8% (95% CI: -36.3%, 3.32%) (p = 0.025). On the other hand, in the population with no intraprocedural deviations, pneumothorax rates were 3.2% and 17.2% in the study and control groups, respectively. The difference between the two groups was -14.0% (95% CI: -29.1%, 1.07%) (p = 0.083). In the control group, 3.45% of the cases required aspiration, while no intervention was required in the study group.

CONCLUSION

The EPABI application along with IABPI is a promising method to decrease the incidence of pneumothorax following CT-guided percutaneous lung biopsy.

Factors Associated with Chest Tube Placement in Blunt Trauma Patients with an Occult Pneumothorax

Background

An occult pneumothorax is identified by computed tomography but not visualized by a plain film chest X-ray. The optimal management remains unclear.

Methods

A retrospective review of an urban level I trauma center's trauma registry was conducted to identify patients with occult pneumothorax over a 2-year period. Factors predictive of chest tube placement were identified using univariate and multivariate logistic regression analysis.

Results

A total of 131 patients were identified, of whom 100 were managed expectantly with an initial period of observation. Ultimately, 42 (32.0%) patients received chest tubes and 89 did not. The patients who received chest tubes had larger pneumothoraces at initial assessment, a higher incidence of rib fractures, and an increased average number of rib fractures, of which significantly more were displaced.

Conclusions

Displaced rib fractures and moderate-sized pneumothoraces are significant factors associated with chest tube placement in a victim of blunt trauma with occult pneumothorax. The optimal timing for the first follow-up chest X-ray remains unclear.

Automated detection of moderate and large pneumothorax on frontal chest X-rays using deep convolutional neural networks: A retrospective study

BACKGROUND

Pneumothorax can precipitate a life-threatening emergency due to lung collapse and respiratory or circulatory distress. Pneumothorax is typically detected on chest X-ray; however, treatment is reliant on timely review of radiographs. Since current imaging volumes may result in long worklists of radiographs awaiting review, an automated method of prioritizing X-rays with pneumothorax may reduce time to treatment. Our objective was to create a large human-annotated dataset of chest X-rays containing pneumothorax and to train deep convolutional networks to screen for potentially emergent moderate or large pneumothorax at the time of image acquisition.

METHODS AND FINDINGS

In all, 13,292 frontal chest X-rays (3,107 with pneumothorax) were visually annotated by radiologists. This dataset was used to train and evaluate multiple network architectures. Images showing large- or moderate-sized pneumothorax were considered positive, and those with trace or no pneumothorax were considered negative. Images showing small pneumothorax were excluded from training. Using an internal validation set (n = 1,993), we selected the 2 top-performing models; these models were then evaluated on a held-out internal test set based on area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and positive predictive value (PPV). The final internal test was performed initially on a subset with small pneumothorax excluded (as in training; n = 1,701), then on the full test set (n = 1,990), with small pneumothorax included as positive. External evaluation was performed using the National Institutes of Health (NIH) ChestX-ray14 set, a public dataset labeled for chest pathology based on text reports. All images labeled with pneumothorax were considered positive, because the NIH set does not classify pneumothorax by size. In internal testing, our "high sensitivity model" produced a sensitivity of 0.84 (95% CI 0.78-0.90), specificity of 0.90 (95% CI 0.89-0.92), and AUC of 0.94 for the test subset with small pneumothorax excluded. Our "high specificity model" showed sensitivity of 0.80 (95% CI 0.72-0.86), specificity of 0.97 (95% CI 0.96-0.98), and AUC of 0.96 for this set. PPVs were 0.45 (95% CI 0.39-0.51) and 0.71 (95% CI 0.63-0.77), respectively. Internal testing on the full set showed expected decreased performance (sensitivity 0.55, specificity 0.90, and AUC 0.82 for high sensitivity model and sensitivity 0.45, specificity 0.97, and AUC 0.86 for high specificity model). External testing using the NIH dataset showed some further performance decline (sensitivity 0.28-0.49, specificity 0.85-0.97, and AUC 0.75 for both). Due to labeling differences between internal and external datasets, these findings represent a preliminary step towards external validation.

CONCLUSIONS

We trained automated classifiers to detect moderate and large pneumothorax in frontal chest X-rays at high levels of performance on held-out test data. These models may provide a high specificity screening solution to detect moderate or large pneumothorax on images collected when human review might be delayed, such as overnight. They are not intended for unsupervised diagnosis of all pneumothoraces, as many small pneumothoraces (and some larger ones) are not detected by the algorithm. Implementation studies are warranted to develop appropriate, effective clinician alerts for the potentially critical finding of pneumothorax, and to assess their impact on reducing time to treatment.

The role of imaging in the assessment of severe asthma

PURPOSE OF REVIEW

The present review aims to summarize the most recent evidence related to imaging and severe asthma, both with regard to advances in imaging research and to their current and potential clinical implications.

RECENT FINDINGS

Recent work in imaging in severe asthma has principally been using computed tomography (CT) and MRI, as well as the integration of the two. Some of the most notable findings include the use of CT imaging biomarkers to create unique clusters of asthmatics, and the use of co-registration to link CT images of airways with regional variation in ventilation in MRI. In addition, temporal studies have shown that some the ventilation defects found using MRI in asthmatics are intermittent and others are persistent, but both are associated with lower lung function.

SUMMARY

The role of imaging in severe asthma currently is primarily in the exclusion of comorbid or other conditions, or in the assessment for complications in the setting of acute decompensation. A rapidly expanding body of literature using CT and MRI suggests that these tools may soon be of utility in the chronic management of the disease.

Diagnosis and treatment of pneumothorax

Many diagnostic and therapeutic options exist for the evaluation and treatment of patients with pneumothorax. Guidelines from US and European professional societies and individual expert opinions differ in the approach to patient care. Advances in diagnostic techniques, such as real-time thoracic ultrasound, have added to the evaluation strategy. It is important for medical trainees and providers to become familiar with techniques utilized worldwide as they may be encountered in clinical practice. We review current evidence, expert recommendations, and compare professional society guidelines discussing the various diagnostic and management options for patients with pneumothorax to assist physicians and trainees involved in the care of hospitalized and outpatient adults who have primary, secondary, and traumatic iatrogenic pneumothorax. Management of traumatic non-iatrogenic pneumothorax is beyond the scope of this article, thus, not reviewed here.

The intrapleural volume threshold for ultrasound detection of pneumothoraces: an experimental study on porcine models

Background

Small pneumothoraxes (PTXs) may not impart an immediate threat to trauma patients after chest injuries. However, the amount of pleural air may increase and become a concern for patients who require positive pressure ventilation or air ambulance transport. Lung ultrasonography (US) is a reliable tool in finding intrapleural air, but the performance characteristics regarding the detection of small PTXs need to be defined. The study aimed to define the volume threshold of intrapleural air when PTXs are accurately diagnosed with US and compare this volume with that for chest x-ray (CXR).

Methods

Air was insufflated into a unilateral pleural catheter in seven incremental steps (10, 25, 50, 100, 200, 350 and 500 mL) in 20 intubated porcine models, followed by a diagnostic evaluation with US and a supine anteroposterior CXR. The sonographers continued the US scanning until the PTXs could be ruled in, based on the pathognomonic US “lung point” sign. The corresponding threshold volume was noted. A senior radiologist interpreted the CXR images.

Results

The mean threshold volume to confirm the diagnosis of PTX using US was 18 mL (standard deviation of 13 mL). Sixty-five percent of the PTXs were already diagnosed at 10 mL of intrapleural air; 25%, at 25 mL; and the last 10%, at 50 mL. At an air volume of 50 mL, the radiologist only identified four out of 20 PTXs in the CXR pictures; i.e., a sensitivity of 20% (95% CI: 7%, 44%). The sensitivity of CXR increased as a function of volume but leveled off at 67%, leaving one-third (1/3) of the PTXs unidentified after 500 mL of insufflated air.

Conclusion

Lung US is very accurate in diagnosing even small amounts of intrapleural air and should be performed by clinicians treating chest trauma patients when PTX is among the differential diagnoses.

A porcine pneumothorax model for teaching ultrasound diagnostics

Objectives: Ultrasound (US) is a sensitive diagnostic tool for detecting pneumothorax (PTX), but methods are needed to optimally teach this technique outside of direct patient care. In training and research settings, porcine PTX models are sometimes used, but the description of the PTX topography in these models is lacking. The study purpose was to define the distribution of air using the reference imaging standard computed tomography (CT), to see if pleural insufflation of air into a live anaesthetized pig truly imitates a PTX in an injured patient.

Methods: A unilateral catheter was inserted into one pleural cavity of each of 20 pigs, and 500 mL of air was insufflated. After a complete thoracic CT scan, the anterior, lateral, medial, basal, apical, and posterior components of the PTXs were compared. The amount of air in each location was quantified by measuring the distance from the lung edge to the chest wall (LE-CW). A supine anteroposterior chest radiograph (CXR) was taken from each model and interpreted by a senior radiologist, and the image results were compared to CT.

Results: All 20 hemithoraces with PTX were correctly identified by CT, while six remained occult after interpreting the CXRs. The PTXs were anterior (100%), lateral (95%), medial (80%), basal (60%), apical (45%), and posterior (15%). The major proportion of the insufflated 500-mL volume was found in the anterior, medial, and basal recesses.

Conclusions: The authors found the distribution of the intrathoracic air to be similar between a porcine model and that to be expected in human trauma patients, all having predominantly anterior PTX topographies. In a training facility, the model is easy to set up and can be scanned by the participants multiple times. To acquire the necessary skills to perform thoracic US examinations for PTX, the porcine models could be useful.

My Patient is Short of Breath: Have They Got a Pneumothorax?

Clinical diagnosis of pneumothorax in the emergency department (ED) resuscitation room can be difficult and in certain circumstances chest radiography is either impractical or the delay is unacceptable. The diagnosis must also be considered in other clinical areas such as critical care, theatres, respiratory units and acute medical units. Erect chest radiography is the standard first-line diagnostic test for pneumothorax in the ED, but the sensitivity is low (59–81%). For many patients in ED, critical care and theatre, an erect chest radiograph is not possible as the patient must remain supine. The sensitivity for a supine chest radiograph has been reported as being 45–75%. Ultrasound has been shown to be more sensitive than chest radiography in the diagnosis of pneumothorax in certain settings. This paper outlines the evidence for ultrasound in the diagnosis of pneumothorax in the point of care setting; it describes the technique and discusses the clinical application of this imaging modality.

Emergency room management of patients with blunt major trauma: evaluation of the multislice computed tomography protocol exemplified by an urban trauma center

BACKGROUND

The early clinical management of patients with major trauma still represents a challenge. To clinically evaluate the full extent of a patient's injuries is difficult, especially when the patient is unconscious. Before December 2002, trauma patients admitted to our emergency room (ER) underwent a diagnostic protocol including physical examination, conventional radiography (CR), sonography and further procedures if necessary. After the installation of a MSCT scanner, all trauma patients underwent the "MSCT protocol" immediately after admission. The aim of the study was to compare the "MSCT-protocol" as it is performed at our institution, with the "Pre-MSCT-protocol".

METHODS

We compared 185 patients undergoing the "Pre MSCT-protocol" with 185 patients undergoing "MSCT protocol". We evaluated the efficacy, speed and accuracy of the "MSCT protocol" using several variables. Time periods from admission to the ER to admission to the intensive care unit were compared as well as outcome parameters such as length of ICU stay, ventilation period and rates of organ. Dichotomous data were analyzed by Chi-square analysis; continuous data were analyzed by Student's t test. Any values of p < 0.05 were considered significant for any test.

RESULTS

No significant differences were found regarding demographic data. The full extent of injuries was definitively diagnosed after 12 +/- 9 minutes in 92.4% of the "MSCT protocol" cohort. In only 76.2% of "Pre-MSCT protocol" cohort definitive diagnosis was possible after 41 +/- 27 minutes. Total ER time was 104 +/- 21 minutes with the "Pre-MSCT protocol" and 70 +/- 17 minutes with "MSCT protocol" (p < 0.05). "Pre-MSCT protocol" patients had a significantly longer ICU stay than "MSCT protocol" patients (p < 0.05). "MSCT protocol" patients had significantly fewer ventilation days (14.3 vs. 10.9 days). Furthermore, rates of organ failure were lower in patients undergoing the "MSCT protocol".

CONCLUSION

We could demonstrate that immediate MSCT in patients with blunt major trauma leads to more accurate and faster diagnosis, and reduction of early clinical time intervals. We also observed a reduction in ventilation, ICU, and hospital days, and in organ failure rates, though this might have been partly due to small differences in case mix. The "MSCT protocol" algorithm seems to be safe and effective.

Imaging of Pleural Disease

Imaging plays an important role in the diagnosis and subsequent management of patients with pleural disease. The presence of a pleural abnormality is usually suggested following a routine chest x-ray, with a number of imaging modalities available for further characterization. This article describes the radiographic and cross-sectional appearances of pleural diseases, which are commonly encountered in every day practice. The conditions covered include benign and malignant pleural thickening, pleural effusions, empyema and pneumothoraces. The relative merits of CT, MRI and PET in the assessment of these conditions and the role of image-guided intervention are discussed.

Chest Pain: A Clinical Assessment

Chest pain is one of the most common presentations in emergency medicine. The initial evaluation should always consider life-threatening causes such as aortic dissection, pulmonary embolism, pneumothorax, pneumomediastinum, pericarditis, and esophageal perforation. Radiographic imaging is performed in tandem with the initial clinical assessment and stabilization of the patient. Radiologic findings are key to diagnosis and management of this entity.

Value of high-resolution ultrasound in detecting a pneumothorax

This study was designed to compare the detectability of high-resolution ultrasound (HRUS) and bedside chest radiography (CR) for a pneumothorax. During the last 14 months, 97 consecutive patients who were admitted to our institute to undergo a transthoracic needle aspiration and biopsy (TNAB) of the lung were included. Both HRUS and CR were performed immediately after the TNAB procedure. The HRUS images were saved in sequence using the cine-review mode and displayed as an animation on a workstation. Four radiologists independently analyzed both HRUS images and a soft copy of the CR on a diagnostic monitor and identified the pneumothorax. With CT as the reference standard, statistical parameters were calculated. From 97 patients, 35 pneumothorax cases were found on CT after the TNAB. The sensitivities in detecting the pneumothorax were 80 and 47% in HRUS and CR, respectively. The specificities were 94 and 94%. The diagnostic accuracies were 89 and 77%. The inter-observer agreement was excellent (kappa=0.85) in the HRUS images and moderate (kappa=0.49) in the CR. The results of this study suggest that HRUS is a more sensitive and confident method for diagnosing a pneumothorax when compared to bedside chest radiography.

Radiology in pleural disease: State of the art

Diseases of the pleura and pleural space are common and present a significant contribution to the workload of the chest radiologist. The radiology department plays a crucial role in the imaging and management of pleural disease. This review aims to describe and illustrate the appearances of common pleural pathologies on various radiological modalities including plain film, ultrasound, CT, magnetic resonance imaging and positron emission tomography. The review will also address the state-of-the-art techniques used to image pleural disease and discuss image-guided intervention in the management of pleural disease.

[Pneumothorax]

During traumatic lesions, lung diseases or even spontaneously air can reach the pleural cavity causing a so called pneumothorax. If a valve mechanism develops, the high pressure in the pleural cavity impaires the blood flow to the heart, causing a circulatory insufficiency. This situation can develop very quickly and therefore must be very fast diagnosed and treated because of the life threatening character. Clinical symptoms and findings should be the main stay in the diagnosis. X-ray and CT may provide additional information concerning the cause of the disease. The therapy consists of draining the air by tube suction or by surgical closing of the leak in the pleura. The paper gives some practical information concerning the therapy.

Detection of porcine bone lesions and fissures: comparing digital selenium, digital luminescence, and analog film-screen radiography

OBJECTIVE

The purpose of our study was to compare the diagnostic performance of a digital selenium detector (Thoravision) with that of analog film-screen systems and digital luminescence radiography in skeletal radiography for the detection of fissures and lesions in porcine bones.

MATERIALS AND METHODS

One hundred bones taken from domestic pigs (50 ribs and 50 femurs) were divided into two equal groups. Fissures and bone lesions were created in 50 bones and 50 served as controls. The bones were examined using film-screen systems, digital luminescence radiography, and digital selenium radiography at various doses. Digital selenium radiography exposure values were adapted to the image geometry differing from the reference methods with a detector focus distance of 2.15 m. Four radiologists independently evaluated image quality and detectability of fissures and lesions on a five-point scale of confidence. Statistical evaluation was based on receiver operating characteristic curve analysis.

RESULTS

Fissures and bone lesions were detected most reliably using the mammography film-screen system, but the difference in the results of the analog and digital reference images did not achieve statistical significance.

CONCLUSION

Compared with analog film-screen systems, the lower spatial resolution of the digital selenium and digital luminescence radiography systems does not affect detectability of fissures and bone lesions in porcine bone. Selenium is effective in skeletal radiography for detecting fissures and bone lesions. With digital selenium and digital luminescence radiography, the surface dose can be cut to half that required for 200-speed film-screen systems without losing any diagnostically relevant information.

Computed radiography

CR is a useful tool for the equine veterinarian and has many advantages. Its only major disadvantage is the initial cost, in dollars and time, for setup in one's practice. CR is already in use at several universities and private practices in the United States and around the world. In the future, as is the case with any computer-based technology, this modality should become more affordable and readily available to smaller practices. The potential of CR in veterinary medicine combined with saturation of the human market is driving the development of specialized software and algorithms for veterinary medicine.

Detection and localization of pulmonary air leaks using laser-polarized (3)He MRI

Pulmonary air leaks were created in the lungs of Yorkshire pigs. Dynamic, 3D MRI of laser-polarized (3)He gas was then performed using a gradient-echo pulse sequence. Coronal magnitude images of the helium distribution were acquired during gas inhalation with a voxel resolution of approximately 1.2 x 2.5 x 8 mm, and a time resolution of 5 sec. In each animal, the ventilation images reveal focal high-signal intensity within the pleural cavity at the site of the air leaks. In addition, a wedge-shaped region of increased parenchymal signal intensity was observed adjacent to the site of the air leak in one animal. (3)He MRI may prove helpful in the management of patients with pulmonary air leaks.

Detection of occult pneumothoraces on abdominal computed tomographic scans in trauma patients

BACKGROUND

An unrecognized, or occult, injury is particularly dangerous in trauma patients, who often have multiple life-threatening injuries. We sought to determine the frequency and quantify the utility of the abdominal computed tomographic (CT) scan in detecting occult pneumothoraces.

METHODS

Charts of all trauma patients diagnosed with a pneumothorax over a 7-year period were retrospectively reviewed for findings on admission chest radiograph, abdominal CT scan, and other demographic data.

RESULTS

A total of 312 charts were reviewed, with 230 patients identified as having a pneumothorax at admission. Of the 230 patients, 126 (54.8%) had an occult pneumothorax identified on abdominal CT scan. In these patients, the mean Trauma Score was 14.1 +/- 2.9 and the mean Injury Severity Score was 24.1 +/- 11.2. In the group of occult pneumothoraces, 84 (66.7%) underwent chest tube placement.

CONCLUSIONS

Abdominal CT scanning provided important information about thoracic trauma often missed on initial evaluation in the trauma bay. This information frequently affected the patient's clinical management.

Diagnosis of pneumothorax in critically ill adults

The diagnosis of pneumothorax is established from the patients' history, physical examination and, where possible, by radiological investigations. Adult respiratory distress syndrome, pneumonia, and trauma are important predictors of pneumothorax, as are various practical procedures including mechanical ventilation, central line insertion, and surgical procedures in the thorax, head, and neck and abdomen. Examination should include an inspection of the ventilator observations and chest drainage systems as well as the patient's cardiovascular and respiratory systems.Radiological diagnosis is normally confined to plain frontal radiographs in the critically ill patient, although lateral images and computed tomography are also important. Situations are described where an abnormal lucency or an apparent lung edge may be confused with a pneumothorax. These may arise from outside the thoracic cavity or from lung abnormalities or abdominal viscera inside the chest.

Validity of CT classification on management of occult pneumothorax: a prospective study

OBJECTIVE

In the setting of blunt trauma, abdominal CT, which routinely includes images of the lower thorax, frequently reveals pneumothoraces that have not been detected on routine supine chest radiographs. Proper management of these occult pneumothoraces remains controversial. The purpose of this study was to test the hypothesis that small (minuscule) to moderate (anterior) radiographically occult pneumothoraces can be safely managed without chest tube placement for patients in whom the need for positive pressure ventilation is not anticipated.

SUBJECTS AND METHODS

We undertook a prospective study in which 44 occult pneumothoraces were classified into three groups, minuscule, anterior, or anterolateral, according to size and location on CT scans. Choice of initial management (tube thoracostomy versus close observation) was based in part on this classification system and in part on individual circumstances of a surgeon's decision.

RESULTS

Of the 44 pneumothoraces found in 36 patients, 16 pneumothoraces were minuscule, 20 were anterior, and eight were anterolateral. Thirteen minuscule pneumothoraces and 11 anterior pneumothoraces initially managed with observation did not require subsequent tube thoracostomy. All eight patients with anterolateral pneumothoraces underwent tube thoracostomy.

CONCLUSION

Most small (minuscule) occult pneumothoraces can successfully be managed with close observation. The risk that the pneumothorax will progress is slight. Moderate-sized (anterior) pneumothoraces may also be successfully managed without initial placement of a chest tube if the patient is not to undergo positive pressure ventilation.

Phosphor-stimulated computed cephalometry: reliability of landmark identification

The aim of this randomized, controlled, prospective study was to determine the reliability of computed lateral cephalometry (Fuji Medical Systems, Tokyo, Japan) in terms of landmark identification compared to conventional lateral cephalometry (CAWO, Schrobenhausen, Germany). To assess the reliability of landmark identification on lateral cephalographs, 20 computed images, taken at 30 per cent reduced radiation (70 kV, 15 mA, 0.35 s) were compared to 20 conventional images (70 kV, 15 mA, 0.5 s). The 40 lateral cephalographs were taken from 20 orthodontic patients at immediate post-treatment and 1 year after retention. The order and type of imaging was randomized. Five orthodontists identified eight skeletal, four dental and five soft tissue landmarks on each of the 40 films. The error of identification was analysed in the XY Cartesian co-ordinate following digitization. Skeletal landmarks exhibited characteristic dispersion with respect to the Cartesian co-ordinates. Root apices were more variable than crown tips. Soft tissue landmarks were more consistent in the X co-ordinate. Two-way ANOVA shows that there is no significant difference between the two imaging systems in both co-ordinates (P > 0.05). Moreover, the differences are generally small (< 0.5 mm), and are unlikely to be of clinical significance. Most of the variables attained statistical power of at least 0.8 in the X-co-ordinate while only the dental landmarks achieved statistical power of at least 0.78 in the Y-co-ordinate. Based on the results of the study: (1) computed lateral cephalographs can be taken at 30 per cent radiation reduction, compared to conventional lateral cephalograph; (2) each anatomical landmark exhibits its characteristic dispersion of error in both the Cartesian co-ordinates; (3) there is no trend between the two imaging systems, with equivocal result, and none of the landmarks attained statistical significance when both raters and imaging systems are considered as factorial variables; (4) the random error of raters in landmark identification after replicate tracing was highlighted and needs to be taken into consideration in all studies involving landmark identification.

Role of CT in the management of pneumothorax in patients with complex cystic lung disease

The diagnosis and treatment of pneumothorax in patients with complex cystic lung disease may be difficult when relying on plain chest radiography alone. We report four cases in which management was greatly facilitated by the use of CT scanning of the chest.

Chest radiography in the ICU

Proper positioning and assessment of abnormalities and complications of the above-mentioned devices have a significant impact on the management of critically ill patients in the intensive care unit (ICU). The timely assessment of new or rapidly evolving findings is critical. Optimal radiographic technique, availability of images to the clinicians, and rapid reporting by the radiologist all serve to maximize the efficacy of bedside chest radiography in the ICU. Sometimes, changes in cardiopulmonary status may only be appreciated on chest radiographs (CXRs). Complications from ventilatory assistance, such as barotrauma, occur frequently and must be detected promptly. The position of monitoring devices, an important component of critical care management, is best checked radiographically. Indications for CXRs and the recommended frequency for repeat follow-up CXRs are based on the existing literature and the consensus of an expert panel formed by the American College of Radiology.

Clinical experiences with computed radiography

Computed radiography (CR) based on photostimulable phosphor is currently the only feasible way for a radiological department to digitize the bulk of radiological data: the lung and skeletal examinations. Regarding the quality of images for diagnostic purposes, CR imaging is never inferior to a screen/film system (SF) and for several clinical entities CR is superior. Of the many processing possibilities of the image plate (IP) image, the unsharp masking or edge enhancement should be used at a minimum. Dose reduction with CR ranges from 15% to 95%; at our institution it is 37%. Softcopy reading of CR images is advantageous due to the many postprocessing and improved display facilities. Currently there is little use for a 4000 x 4000 (4 K) pixel imaging and display. All images (including mammography) can be read in 2 K without any loss of clinically important information. To include CR in a picture archive and communication system (PACS) is demanding because of the load of data that each CR image represents. Networks for image distribution are essential if digital imaging is to have any impact on patient treatment and hospital organization.

Traumatic chest lesions in patients with severe head trauma: a comparative study with computed tomography and conventional chest roentgenograms

In patients with severe craniocerebral trauma, who need a continuous positive-pressure breathing, the detection of pulmonary and mediastinal traumatic lesions, especially pneumothorax, may alter the management. The aim of this study is to evaluate the efficiency and accuracy of conventional supine chest roentgenograms to detect the associated traumatic chest lesions in severe craniocerebral trauma and to compare their value as a diagnostic method for the identification of unsuspected lesions with a limited chest computed tomographic (CT) examination. Forty-seven consecutive patients with severe craniocerebral trauma underwent head CT and a prospective limited CT examination of the thorax in the same session. Nine patients (19.1%) presented a pneumothorax, bilateral in one case. Six pneumothoraces (60%) were identified both on conventional chest roentgenograms and CT, whereas in four cases (40%), the lesion was only detectable on CT. The CT study also showed 31 areas of pulmonary parenchymal contusions in 19 subjects (40%), whereas the conventional chest roentgenograms demonstrated 17 areas of contusions in 11 (23%) subjects. One thoracic aorta and one right diaphragm rupture were detected on CT study. On the conventional chest roentgenograms the mediastinal vascular injury was overlooked, whereas the right diaphragmatic rupture was highly suspected. The limited chest CT examination supplied additional information in 30% of patients. In 12.7% of patients, this information was clinically significant enough to alter the management. In patients with severe craniocerebral trauma evaluation of associated chest trauma by a supplementary limited chest CT, examination provides more and precise information about the size and severity of mediastinal and pulmonary lesions with a superior detectability of pneumothorax.

Delayed tension pneumothorax complicating central venous catheterization and positive pressure ventilation

Pneumothorax is the most common complication after central venous catheterization. The diagnosis of pneumothorax may be delayed for hours or days, in some instances because of minimal clinical symptoms or radiographic signs on initial evaluation, or in other instances because of late presentation precipitated by positive pressure ventilation. A case is presented in which a patient developed a tension pneumothorax while under general anesthesia 10 days after central venous line placement. A review of the literature suggests that delayed pneumothorax has an incidence of approximately 0.4% of all central venous access attempts, is much more common after subclavian than internal jugular approaches, especially in difficult or multiple attempts, is asymptomatic in 22%, and results in tension pneumothorax in 22%. End-expiratory upright chest radiographs, the optimal radiographic technique for detection of small pneumothoraces, were obtained in only 19% of reviewed cases. Supine views, the least sensitive radiographic technique, should be carefully reviewed for evidence of basilar hyperlucency, a deep sulcus sign, or a double diaphragm sign. In patients unable to tolerate the upright position, supine views should be supplemented with lateral decubitus, oblique, or cross-table lateral views. Emergency physicians should be aware of the possibility of delayed pneumothorax, as well as optimal radiographic technique for demonstration of small pneumothoraces, and subtle radiographic findings in supine or semirecumbent patients.

Digital chest radiography: clinical aspects

Technical development puts the completely digital radiology department within reach. Presently available systems have slightly inferior spatial resolution than film/screen systems. This is well compensated for by image processing. In the future, the digital radiology department working environment will be different. Systems should be designed that conform as much as possible to radiologists' present way of working to facilitate transition to the new system and avoid unnecessary stress.

Double-contrast barium enema study with computed radiography: assessment in detection of colorectal polyps

Feasibility and diagnostic performance of digital radiography in gastrointestinal imaging have not been fully investigated. We performed double-contrast barium enema study with computed radiography (CR-DCBE) on a routine basis, and the efficacy of this system for detection of colorectal polyps was investigated. The files of 76 patients, who had undergone both CR-DCBE and colonoscopy, were reviewed by two observers who were blinded to the colonoscopic findings. The radiation dose of CR-DCBE was reduced to 50% of conventional film-screen system. By using colonoscopic findings as a reference, the sensitivity and positive predictive value for colorectal polyps were 66% to 71% and 32% to 41%, respectively. There was no significant interobserver difference. The use of postprocessing with gray-scale reversal and edge enhancement did not significantly improve the results. The sensitivity for polyps smaller than 1 cm was comparable with published data of conventional film-screen systems. Our preliminary results show that CR-DCBE has an acceptable sensitivity for detection of colorectal polyps. It is suggested that the use of CR is a promising approach to digital gastrointestinal radiography.

Detection of pneumothorax with lateral shoot-through digital radiography

The aim of this study was to compare the performance of digital frontal radiographs vs digital lateral shoot-through radiographs for the diagnosis of pneumothoraces in supine patients. A total of 146 pairs of radiographs were performed on 32 ventilated patients on the Intensive Care Unit. Each radiograph was independently assessed by three observers for the presence or absence of a pneumothorax. A degree of confidence was assigned to each observation and an image quality score was given to each radiograph. At least two out of three observers positively diagnosed a pneumothorax in 13/146 (8.9%) of the frontal radiographs compared to 43/146 (29.4%) of the lateral radiographs (P < 0.0001), but suboptimal images were obtained more often with lateral shoot-through radiographs than with frontal radiographs. We conclude that digital lateral shoot-through radiographs are significantly more sensitive than digital frontal radiographs for the diagnosis of pneumothoraces in supine patients.