Identification of Idiopathic Pulmonary Fibrosis and Prediction of Disease Severity via Machine Learning Analysis of Comprehensive Metabolic Panel and Complete Blood Count Data.

BACKGROUND: Diagnosis of idiopathic pulmonary fibrosis (IPF) typically relies on high-resolution computed tomography imaging (HRCT) or histopathology, while monitoring disease severity is done via frequent pulmonary function testing (PFT). More reliable and convenient methods of diagnosing fibrotic interstitial lung disease (ILD) type and monitoring severity would allow for early identification and enhance current therapeutic interventions. This study tested the hypothesis that a machine learning (ML) ensemble analysis of comprehensive metabolic panel (CMP) and complete blood count (CBC) data can accurately distinguish IPF from connective tissue disease ILD (CTD-ILD) and predict disease severity as seen with PFT. METHODS: Outpatient data with diagnosis of IPF or CTD-ILD (n = 103 visits by 53 patients) were analyzed via ML methodology to evaluate (1) IPF vs CTD-ILD diagnosis; (2) %predicted Diffusing Capacity of Lung for Carbon Monoxide (DLCO) moderate or mild vs severe; (3) %predicted Forced Vital Capacity (FVC) moderate or mild vs severe; and (4) %predicted FVC mild vs moderate or severe. RESULTS: ML methodology identified IPF from CTD-ILD with AUCTEST = 0.893, while PFT was classified as DLCO moderate or mild vs severe with AUCTEST = 0.749, FVC moderate or mild vs severe with AUCTEST = 0.741, and FVC mild vs moderate or severe with AUCTEST = 0.739. Key features included albumin, alanine transaminase, %lymphocytes, hemoglobin, %eosinophils, white blood cell count, %monocytes, and %neutrophils. CONCLUSION: Analysis of CMP and CBC data via proposed ML methodology offers the potential to distinguish IPF from CTD-ILD and predict severity on associated PFT with accuracy that meets or exceeds current clinical practice.

Circulating metabolic profile in idiopathic pulmonary fibrosis: data from the IPF-PRO Registry.

BACKGROUND: The circulating metabolome, reflecting underlying cellular processes and disease biology, has not been fully characterized in patients with idiopathic pulmonary fibrosis (IPF). We evaluated whether circulating levels of metabolites correlate with the presence of IPF, with the severity of IPF, or with the risk of clinically relevant outcomes among patients with IPF. METHODS: We analyzed enrollment plasma samples from 300 patients with IPF in the IPF-PRO Registry and 100 individuals without known lung disease using a set of targeted metabolomics and clinical analyte modules. Linear regression was used to compare metabolite and clinical analyte levels between patients with IPF and controls and to determine associations between metabolite levels and measures of disease severity in patients with IPF. Unadjusted and adjusted univariable Cox regression models were used to evaluate associations between circulating metabolites and the risk of mortality or disease progression among patients with IPF. RESULTS: Levels of 64 metabolites and 5 clinical analytes were significantly different between patients with IPF and controls. Among analytes with greatest differences were non-esterified fatty acids, multiple long-chain acylcarnitines, and select ceramides, levels of which were higher among patients with IPF versus controls. Levels of the branched-chain amino acids valine and leucine/isoleucine were inversely correlated with measures of disease severity. After adjusting for clinical factors known to influence outcomes, higher levels of the acylcarnitine C:16-OH/C:14-DC were associated with all-cause mortality, lower levels of the acylcarnitine C16:1-OH/C14:1DC were associated with all-cause mortality, respiratory death, and respiratory death or lung transplant, and higher levels of the sphingomyelin d43:2 were associated with the risk of respiratory death or lung transplantation. CONCLUSIONS: IPF has a distinct circulating metabolic profile characterized by increased levels of non-esterified fatty acids, long-chain acylcarnitines, and ceramides, which may suggest a more catabolic environment that enhances lipid mobilization and metabolism. We identified select metabolites that were highly correlated with measures of disease severity or the risk of disease progression and that may be developed further as biomarkers. TRIAL REGISTRATION: ClinicalTrials.gov; No: NCT01915511; URL: www. CLINICALTRIALS: gov .

Disease diagnosis and severity classification in pulmonary fibrosis using carbonyl volatile organic compounds in exhaled breath.

BACKGROUND: Pathophysiological conditions underlying pulmonary fibrosis remain poorly understood. Exhaled breath volatile organic compounds (VOCs) have shown promise for lung disease diagnosis and classification. In particular, carbonyls are a byproduct of oxidative stress, associated with fibrosis in the lungs. To explore the potential of exhaled carbonyl VOCs to reflect underlying pathophysiological conditions in pulmonary fibrosis, this proof-of-concept study tested the hypothesis that volatile and low abundance carbonyl compounds could be linked to diagnosis and associated disease severity. METHODS: Exhaled breath samples were collected from outpatients with a diagnosis of Idiopathic Pulmonary Fibrosis (IPF) or Connective Tissue related Interstitial Lung Disease (CTD-ILD) with stable lung function for 3 months before enrollment, as measured by pulmonary function testing (PFT) DLCO (%), FVC (%) and FEV1 (%). A novel microreactor was used to capture carbonyl compounds in the breath as direct output products. A machine learning workflow was implemented with the captured carbonyl compounds as input features for classification of diagnosis and disease severity based on PFT (DLCO and FVC normal/mild vs. moderate/severe; FEV1 normal/mild/moderate vs. moderately severe/severe). RESULTS: The proposed approach classified diagnosis with AUROC=0.877 ± 0.047 in the validation subsets. The AUROC was 0.820 ± 0.064, 0.898 ± 0.040, and 0.873 ± 0.051 for disease severity based on DLCO, FEV1, and FVC measurements, respectively. Eleven key carbonyl VOCs were identified with the potential to differentiate diagnosis and to classify severity. CONCLUSIONS: Exhaled breath carbonyl compounds can be linked to pulmonary function and fibrotic ILD diagnosis, moving towards improved pathophysiological understanding of pulmonary fibrosis.

Evaluation of stacked ensemble model performance to predict clinical outcomes: A COVID-19 study.

BACKGROUND: The application of machine learning (ML) to analyze clinical data with the goal to predict patient outcomes has garnered increasing attention. Ensemble learning has been used in conjunction with ML to improve predictive performance. Although stacked generalization (stacking), a type of heterogeneous ensemble of ML models, has emerged in clinical data analysis, it remains unclear how to define the best model combinations for strong predictive performance. This study develops a methodology to evaluate the performance of "base" learner models and their optimized combination using "meta" learner models in stacked ensembles to accurately assess performance in the context of clinical outcomes. METHODS: De-identified COVID-19 data was obtained from the University of Louisville Hospital, where a retrospective chart review was performed from March 2020 to November 2021. Three differently-sized subsets using features from the overall dataset were chosen to train and evaluate ensemble classification performance. The number of base learners chosen from several algorithm families coupled with a complementary meta learner was varied from a minimum of 2 to a maximum of 8. Predictive performance of these combinations was evaluated in terms of mortality and severe cardiac event outcomes using area-under-the-receiver-operating-characteristic (AUROC), F1, balanced accuracy, and kappa. RESULTS: The results highlight the potential to accurately predict clinical outcomes, such as severe cardiac events with COVID-19, from routinely acquired in-hospital patient data. Meta learners Generalized Linear Model (GLM), Multi-Layer Perceptron (MLP), and Partial Least Squares (PLS) had the highest AUROC for both outcomes, while K-Nearest Neighbors (KNN) had the lowest. Performance trended lower in the training set as the number of features increased, and exhibited less variance in both training and validation across all feature subsets as the number of base learners increased. CONCLUSION: This study offers a methodology to robustly evaluate ensemble ML performance when analyzing clinical data.

Detection and Monitoring of Interstitial Lung Disease in Patients with Systemic Sclerosis.

PURPOSE OF REVIEW: Interstitial lung disease (ILD) is a common manifestation of systemic sclerosis (SSc). We explore the importance of early detection, monitoring, and management of SSc-ILD. RECENT FINDINGS: All patients with SSc are at risk of ILD and should be screened for ILD at diagnosis using a high-resolution computed tomography (HRCT) scan. Some patients with SSc-ILD develop a progressive phenotype characterized by worsening fibrosis on HRCT, decline in lung function, and early mortality. To evaluate progression and inform treatment decisions, regular monitoring is important and should include pulmonary function testing, evaluation of symptoms and quality of life, and, where indicated, repeat HRCT. Multidisciplinary discussion enables comprehensive evaluation of the available information and its implications for management. The first-line treatment for SSc-ILD is usually immunosuppression. The antifibrotic drug nintedanib has been approved for slowing lung function decline in patients with SSc-ILD. Optimal management of patients with SSc-ILD requires a multidisciplinary and patient-centered approach.

Barriers to nutrition therapy in the critically ill patient with COVID-19.

BACKGROUND: Coronavirus disease 2019 (COVID-19) has created challenges for intensivists, as high ventilatory demands and prolonged hypermetabolism make it difficult to sustain nutrition status. The purpose of this survey was to determine current practices in nutrition therapy and identify barriers to its delivery. METHODS: A survey about delivering nutrition therapy to critically ill patients with COVID-19 was sent to clinicians at academic and community hospitals from September to December 2020. RESULTS: Of 440 who viewed the survey, 199 (45%) completed the questionnaire. Respondents were composed of 30%, physicians and 70% registered dietitians, with 51% representing community programs, 43% academic institutions, and 6% Veterans Affairs centers. Half (49%) had protocols for managing critically ill patients with COVID-19, and 21% had a protocol for nutrition therapy. Although most respondents (83%) attempted to feed by the intragastric route, only 9% indicated that energy/protein needs were met. The biggest barriers to delivery of enteral nutrition (EN) involved the patients unpredictable clinical course and fear of aspiration given the lack of respiratory reserve. Intensivists were reluctant to add supplemental parenteral nutrition (PN) because of perceived lack of benefit. CONCLUSION: The survey results would suggest that strategies for nutrition therapy based on the intragastric infusion of EN are unsuccessful in meeting the energy/protein needs of critically ill patients with COVID-19. It is likely these barriers exist in providing nutrition to non-Covid-19 critically ill patients. Intensivists need protocols that optimally deliver intragastric EN, consider early postpyloric infusion, and address adding supplemental PN in a deteriorating nutrition status.

How Differences in the Disease Process of the COVID-19 Pandemic Pose Challenges to the Delivery of Critical Care Nutrition.

PURPOSE OF REVIEW: The COVID-19 pandemic is a unique disease process that has caused unprecedented challenges for intensive care specialists. The hyperinflammatory hypermetabolic nature of the disease and the complexity of its management create barriers to the delivery of nutritional therapy. This review identifies the key differences which characterize this pandemic from other disease processes in critical illness and discusses alternative strategies to enhance success of nutritional support. RECENT FINDINGS: Prolonged hyperinflammation, unlike any previously described pattern of response to injury, causes metabolic perturbations and deterioration of nutritional status. High ventilatory demands, hypercoagulation with the risk of bowel ischemia, and threat of aspiration in patients with little or no pulmonary reserve, thwart initial efforts to provide early enteral nutrition (EN). The obesity paradox is invalidated, tolerance of EN is limited, intensivists are reluctant to add supplemental parenteral nutrition (PN), and efforts to give sufficient nutritional therapy remain a low priority. The nature of the disease and difficulties providing traditional critical care nutrition lead to dramatic deterioration of nutritional status. Institutions should not rely on insufficient gastric feeding alone but focus instead on redoubling efforts to provide postpyloric deep duodenal/jejunal EN or re-examine the role of supplemental PN in this population of patients with such severe critical illness.

Acute Respiratory Distress Syndrome and Time to Weaning Off the Invasive Mechanical Ventilator among Patients with COVID-19 Pneumonia.

Acute respiratory distress syndrome (ARDS) due to coronavirus disease 2019 (COVID-19) pneumonia is the main cause of the pandemic's death toll. The assessment of ARDS and time on invasive mechanical ventilation (IMV) could enhance the characterization of outcomes and management of this condition. This is a city-wide retrospective study of hospitalized patients with COVID-19 pneumonia from 5 March 2020 to 30 June 2020. Patients with critical illness were compared with those with non-critical illness. We examined the severity of ARDS and other factors associated with (i) weaning patients off IMV and (ii) mortality in a city-wide study in Louisville, KY. Of 522 patients with COVID-19 pneumonia, 219 (41.9%) were critically ill. Among critically ill patients, the median age was 60 years; 53% were male, 55% were White and 32% were African American. Of all critically ill patients, 52% had ARDS, and 38% of these had severe ARDS. Of the 25% of patients who were weaned off IMV, those with severe ARDS were weaned within eleven days versus five days for those without severe ARDS, p = 0.023. The overall mortality for critically ill patients was 22% versus 1% for those not critically ill. Furthermore, the 14-day mortality was 31% for patients with severe ARDS and 12% for patients without severe ARDS, p = 0.019. Patients with severe ARDS versus non-severe ARDS needed twice as long to wean off IMV (eleven versus five days) and had double the 14-day mortality of patients without severe ARDS.

The Role of Surgical Lung Biopsy in the Diagnosis of Fibrotic Interstitial Lung Disease: Perspective from the Pulmonary Fibrosis Foundation.

Diagnosis of interstitial lung disease (ILD) requires a multidisciplinary discussion approach that includes clinicians, radiologists, and pathologists. Surgical lung biopsy (SLB) is currently the recommended standard in obtaining pathologic specimens for patients with ILD requiring a tissue diagnosis. The increased diagnostic confidence and accuracy provided by microscopic pathology assessment of SLB specimens must be balanced with the associated risks in patients with ILD. This document was developed by the SLB Working Group of the Pulmonary Fibrosis Foundation, composed of a multidisciplinary group of ILD physicians, including pulmonologists, radiologists, pathologists, and thoracic surgeons. In this document, we present an up-to-date literature review of the indications, contraindications, risks, and alternatives to SLB in the diagnosis of fibrotic ILD; outline an integrated approach to the decision-making around SLB in the diagnosis of fibrotic ILD; and provide practical information to maximize the yield and safety of SLB.

Intracranial hemorrhage in a young COVID-19 patient.

COVID-19 patients are increasingly understood to develop multisystem manifestations, including neurologic involvement. We report the case of a 42-year old COVID-19 positive patient with a fatal intracerebral hemorrhage (ICH). The patient presented with fever and dyspnea, requiring intubation due to medical complications. After prolonged sedation and anticoagulation, the patient suddenly developed bilaterally fixed and dilated pupils, caused by a right-sided intracranial hemorrhage with uncal herniation. The course of this case illustrates the delicate balance between hypercoagulability and coagulation factor depletion; especially in the intubated and sedated patient, in whom regular neurological assessments are impeded. As we expand our understanding of the neurological ramifications of COVID-19, clinicians need to be increasingly aware of the precarious coagulation balance.

Scleroderma-related interstitial lung disease.

Scleroderma-related interstitial lung disease (SSc-ILD) is a pulmonary fibrosing disorder characterized by systemic inflammation and progressive scarring of the lungs that leads to respiratory failure. Although certain immunosuppressive therapies may slow disease progression, current treatment strategies are not curative; consequently, SSc-ILD continues to be a major cause of morbidity and mortality. We present four cases of SSc-ILD that emphasize the importance of early screening and detection, close follow-up, and aggressive management. We also highlight the need for well-conducted clinical trials designed to identify new and effective treatments.