Deep learning model to predict the need for mechanical ventilation using chest X-ray images in hospitalised patients with COVID-19.

OBJECTIVES: There exists a wide gap in the availability of mechanical ventilator devices and their acute need in the context of the COVID-19 pandemic. An initial triaging method that accurately identifies the need for mechanical ventilation in hospitalised patients with COVID-19 is needed. We aimed to investigate if a potentially deteriorating clinical course in hospitalised patients with COVID-19 can be detected using all X-ray images taken during hospitalisation. METHODS: We exploited the well-established DenseNet121 deep learning architecture for this purpose on 663 X-ray images acquired from 528 hospitalised patients with COVID-19. Two Pulmonary and Critical Care experts blindly and independently evaluated the same X-ray images for the purpose of validation. RESULTS: We found that our deep learning model predicted the need for mechanical ventilation with a high accuracy, sensitivity and specificity (90.06%, 86.34% and 84.38%, respectively). This prediction was done approximately 3 days ahead of the actual intubation event. Our model also outperformed two Pulmonary and Critical Care experts who evaluated the same X-ray images and provided an incremental accuracy of 7.24%-13.25%. CONCLUSIONS: Our deep learning model accurately predicted the need for mechanical ventilation early during hospitalisation of patients with COVID-19. Until effective preventive or treatment measures become widely available for patients with COVID-19, prognostic stratification as provided by our model is likely to be highly valuable.

Analysis of the 30-Pack-Year Smoking Threshold in African Americans From an Underserved Lung Cancer Screening Program.

PURPOSE: African Americans were underrepresented in lung cancer screening (LCS) trials, despite having higher lung cancer incidence and worse outcomes compared with Caucasians. There is concern that the 30-pack-year threshold excludes some African Americans who may benefit from LCS. METHODS: LCS in an underserved health care system was reviewed. Providers attested that patients met LCS criteria, including 30-pack-year history, but patients also self-reported smoking histories. Self-reported data were used to identify patients with <30-pack-year histories. RESULTS: Over 2 years, 784 patients self-reported sufficient data to calculate pack-years. The majority were men (57.5%), and 66.2% were African Americans. Median total years smoked was 40 (interquartile range, 30-45 years), and median pack-years was 25 (interquartile range, 15-40 pack-years). African Americans were more likely to report <30 pack-years compared with other races (P < .001). The overall incidence of lung cancer was 2.0%, and incidence was similar for those with ≥30 or <30 pack-years (2.1% versus 2.0%; odds ratio, 0.94; 95% confidence interval, 0.35-2.53; P = .902). Race was not associated with lung cancer diagnosis, but African Americans were the only race to have lung cancer if pack-years were <30. The incidence of cancer in African Americans was similar in those who reported ≥30 or <30 pack-years (2.2% versus 2.7%; odds ratio, 1.21; 95% confidence interval, 0.39-3.75; P = .740), and the 30-pack-year threshold was not associated with lung cancer diagnosis. CONCLUSIONS: This is the first review of LCS in African Americans who self-reported <30 pack-years. Although retrospective, these data raise concern that the 30-pack-year threshold may not be an appropriate LCS criterion in African Americans.

Acute respiratory failure in randomized trials of noninvasive respiratory support: A systematic review of definitions, patient characteristics, and criteria for intubation.

PURPOSE: To examine the definitions of acute respiratory failure, the characteristics of recruited patients, and the criteria for intubation used in randomized trials. METHODS: We searched MEDLINE for randomized trials of noninvasive respiratory support modalities in patients with de novo respiratory failure. We included trials from 1995 to 2017 that enrolled 40 or more patients and used intubation as an outcome. RESULTS: We examined the reports of 53 trials that enrolled 7225 patients. There was wide variation in the use of variables for defining acute respiratory failure. Dyspnea was rarely measured and the increase in breathing effort was poorly defined. The characteristics of patients enrolled in trials changed over time and differed by the cause of respiratory failure. Intubation was poorly characterized. The criteria for intubation had more variables than the criteria for respiratory failure. CONCLUSIONS: We identified deficiencies in the design and reporting of randomized trials, some of which can be remedied by investigators. We also found that patient characteristics differ by the type of respiratory failure. This knowledge can help clinician identify patients at the right moment to benefit from the tested interventions and investigators in developing criteria for enrollment in future trials.

Extracorporeal membrane oxygenation in severe influenza infection with respiratory failure: A systematic review and meta-analysis.

INTRODUCTION: Extracorporeal membrane oxygenation (ECMO) has been extensively used for potentially reversible acute respiratory failure associated with severe influenza A (H1N1) pneumonia; however, it remains an expensive, resource-intensive therapy, with a high associated mortality. This systematic review and meta-analysis aims to summarize and pool outcomes data available in the published literature to guide clinical decision-making and further research. METHODS: We conducted a systematic search of MEDLINE (1966 to April 15, 2015), EMBASE (1980 to April 15, 2015), CENTRAL, and Google Scholar for patients with severe H1N1 pneumonia and respiratory failure who received ECMO. The study validity was appraised by Newcastle-Ottawa Scale. The primary outcome was all-cause mortality. The secondary outcomes were duration of ECMO therapy, mechanical ventilation, and Intensive Care Unit (ICU) length of stay. RESULTS: Of 698 abstracts screened and 142 full-text articles reviewed, we included 13 studies with a total of 494 patients receiving ECMO in our final review and meta-analysis. The study validity was satisfactory. The overall mortality was 37.1% (95% confidence interval: 30-45%) limited by underlying heterogeneity (I2 = 65%, P value of Q statistic = 0.006). The median duration for ECMO was 10 days, mechanical ventilation was 19 days, and ICU length of stay was 33 days. Exploratory meta-regression did not identify any statistically significant moderator of mortality (P < 0.05), except for the duration of pre-ECMO mechanical ventilation in days (coefficient 0.19, standard error: 0.09, Z = 2.01, P < 0.04, R2 = 0.16). The visual inspection of funnel plots did not suggest the presence of publication bias. CONCLUSIONS: ECMO therapy may be used as an adjunct or salvage therapy for severe H1N1 pneumonia with respiratory failure. It is associated with a prolonged duration of ventilator support, ICU length of stay, and high mortality. Initiating ECMO early once the patient has been instituted on mechanical ventilation may result in improved survival.

Effect of Wakefulness-Promoting Agents on Sleepiness in Patients with Sleep Apnea Treated with CPAP: A Meta-Analysis.

STUDY OBJECTIVES: To perform a meta-analysis of the effect of wakefulness-promoting agents (modafinil and armodafinil) in patients with residual sleepiness after CPAP therapy for obstructive sleep apnea. METHODS: We conducted a systematic search of MEDLINE (1966 to September 2014), EMBASE (1980 to September 2014) and Cochrane Database for randomized placebo controlled trials on modafinil or armodafinil in patients who met established criteria for diagnosis of obstructive sleep apnea, adequate continuous positive airway pressure use, and who complained of residual sleepiness. Risk of bias was assessed. Primary outcomes were the Epworth Sleepiness Scale and mean sleep latencies on the maintenance of wakefulness test. Secondary outcomes were the Clinical Global Impression of Change, change in daily continuous positive airway pressure use, and the frequency of headaches. RESULTS: Out of 118 abstracts screened and 12 full text articles reviewed, we included 6 studies (total of 1,479 participants) in our final meta-analysis: Three evaluated modafinil, and three armodafinil. Risk of bias was unclear in one or more key domains for four studies. When compared with placebo, wakefulness promoting agents decreased Epworth Sleepiness Scale by 2.51 points (95% CI, 2.00-3.02), increased sleep latency in maintenance of wakefulness test by 2.73 minutes (95% CI, 2.12-3.34), increased the reporting of minimal improvement on the Clinical Global Impression of Change by 26% (RR 1.59; 95% CI, 1.36-1.86), and increased the risk of headaches by 8% (RR 1.98; 95% CI, 1.48-2.63). Also, there was a trend for decreased continuous positive airway pressure after treatment with these agents. CONCLUSION: Wakefulness promoting agents improve objective and subjective measures of sleepiness, wakefulness, perception of disease severity in patients with residual sleepiness after CPAP therapy for OSA, and are generally well tolerated.

Aorto-right atrial fistula: Late complication of tricuspid valve infective endocarditis.

Abnormal connections between the ascending aorta and the cardiac chambers are rare, especially in the context of right-sided infective endocarditis (IE). Transthoracic echocardiography (TTE) with color-flow Doppler, transesophageal echocardiography (TEE), or both may be required for diagnosis. We present the case of a woman admitted with right-sided heart failure (HF) symptoms. She had a previous history of tricuspid valve IE 30 years ago. TTE and TEE revealed an aorto-right atrium fistula located just under the non-coronary cusp into the right atrium at the level of the previously affected tricuspid valve. The Patient refused surgery and was discharged home on HF medications. She has been stable for the last 3 years. The peculiarity of this case is the late symptomatic presentation of the aorto-atrial fistula and the unusual association to tricuspid valve IE.

Drug-induced immune haemolytic anaemia caused by levofloxacin.

Drug-induced immune haemolytic anaemia (DIIHA) is extremely rare. We herein report a case of life-threatening DIIHA due to levofloxacin. This is the second case reported in the literature. A 51-year-old woman presented with complaints of fatigue after 4-5 days of levofloxacin therapy for a lung infection. At presentation, she was found to have haemolysis with a positive Coombs test and IgG autoantibodies. Levofloxacin was identified as the probable culprit, using the Naranjo adverse drug reaction probability scale. Upon discontinuation of the drug and initiation of steroids, the patient's haematological parameters stabilised. Diagnosis of DIIHA is made through a history of intake of levofloxacin, clinical and laboratory features of haemolysis and a positive Coombs test. An autoantibody screen is most commonly positive for warm antibodies (IgG type). It is essential for clinicians to recognise this rare complication caused by a commonly prescribed medication, discontinue the offending drug and initiate treatment.

Quick diagnosis units--an effective alternative to hospitalization for diagnostic workup: a systematic review.

INTRODUCTION: This era of healthcare reform emphasizes improving value of care. Inpatient admissions for diagnostic evaluation put economic pressure on an already strained healthcare system. We conducted a systematic review of effectiveness of quick diagnosis units (QDUs), an established outpatient model for early diagnostic workups in Europe. METHODS: We searched MEDLINE and Embase for studies that focused on implementation of quick/rapid diagnosis units, with relevant Medical Subject Headings terms and keywords. Of 2047 studies, we selected 13 for full-text screening and bibliography review. Of these, 5 studies included at least 2 primary outcomes of interest and were included in our review. These units functioned as outpatient clinics, staffed by internists, nurses, and clerical staff, with expedited scheduling of outpatient diagnostic tests. Our primary outcome measures were final diagnosis, the mean time to final diagnosis, inpatient bed-days saved per patient, and costs saved per patient. We also studied discharge disposition, care preferences, and safety data. RESULTS: About 18% to 30% of patients were diagnosed with malignancy, with an average time to diagnosis of 6 to 11 days. Inpatient bed-days saved per patient ranged from 4.5 to 7. Savings from fixed costs of hospitalization ranged from $2336(€1764) to $3304(€2514) for each patient enrolled in the protocol. The QDU model was preferred by 88% of patients, and patient satisfaction rates were 95% to 97%. CONCLUSIONS: QDUs seem an effective and cost-saving alternative to inpatient hospitalization, and appear to be a safe approach for diagnostic workup of potentially severe diseases in select patient populations, although there are limited safety data available.

Acute gastritis-induced Takotsubo's cardiomyopathy.

KEY CLINICAL MESSAGE: A 50-year-old lady presented with epigastralgia, electrocardiogram (ECG) showed T-wave inversions and the echocardiogram low ejection fraction (EF) with apical ballooning. An esophagogastroduodenoscopy (EGD) revealed gastritis. She recovered with proton pump inhibitors treatment. This is the first case that describes gastritis-induced stress cardiomyopathy. Clinicians should be aware of Takotsubo's cardiomyopathy (TCM) as a possible complication of gastritis.