Dataset Information

A risk-predictive model for invasive pulmonary aspergillosis in patients with acute exacerbation of chronic obstructive pulmonary disease.

ABSTRACT:

Objectives

Invasive pulmonary aspergillosis (IPA) is increasingly reported in chronic obstructive pulmonary disease (COPD) patients. These patients often have poor clinical outcomes. Early recognition of IPA in COPD is always challenging. We aimed to develop and validate a risk model using readily available clinical parameters to predict IPA for acute exacerbation of COPD (AECOPD) patients.

Methods

We performed a retrospective cohort study. AECOPD patients who were admitted to Jinling Hospital between January 2012 and December 2017 were included. 880 AECOPD patients were randomly divided into the training set (70%, n = 616) and validation set (30%, n = 264). A nomogram model was developed using multivariate logistic regression from training set. The discrimination and calibration of model were validated internally. Decision curve analyses assessed the clinical utility of the nomogram.

Results

The incidence of IPA in hospitalized AECOPD patients was 9.6% in the training set (59 cases of IPA) and 9.1% in the validation set (24 cases of IPA), respectively. The nomogram model consisted of independent factors associated with IPA included lung function GOLD III-IV, use of broad-spectrum antibiotic over 10 days in the last month, oral or intravenous corticosteroids (prednisone) over 265 mg in the last 3 months and serum albumin < 30 g/L. The model performed good discrimination and calibration in validation set (c-statistic, 0.79 [95%CI 0.68-0.90]). The 95%CI region of calibration belt did not cross the 45-degree diagonal bisector line (P = 0.887).

Conclusion

The simple risk predictive model for earlier recognition of IPA is useful in hospitalized AECOPD patients.

SUBMITTER: Gu Y

PROVIDER: S-EPMC8188951 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundAcute exacerbation of chronic obstructive pulmonary disease (AECOPD) is a clinical syndrome with various causes. It is not uncommon that COPD patients presenting with dyspnea have multiple causes for their symptoms including AECOPD, pneumonia, or congestive heart failure occurring concurrently.MethodsTo identify clinical, radiographic, and laboratory characteristics that might help distinguish AECOPD from another dominant disease in patients with a history of COPD, we conducted a retrospective cohort study of hospitalized patients with admitting diagnosis of AECOPD who were screened for a prospective randomized controlled trial from Sep 2016 to Mar 2018. Clinical characteristics, course in hospital, and final diagnosis at discharge were reviewed and adjudicated by two authors. The final diagnosis of each patient was determined based on the synthesis of all presenting signs and symptoms, imaging, and laboratory results. We adhered to AECOPD diagnosis definitions based on the GOLD guidelines. Univariate and multivariate analyses were performed to identify any associated features of AECOPD with and without other acute processes contributing to dyspnea.ResultsThree hundred fifteen hospitalized patients with admitting diagnosis of AECOPD were included. Mean age was 72.5 (SD 10.6) years. Two thirds (65.4%) had spirometry defined COPD. The most common presenting symptom was dyspnea (96.5%), followed by cough (67.9%), and increased sputum (57.5%). One hundred and eighty (57.1%) had a final diagnosis of AECOPD alone whereas 87 (27.6%) had AECOPD with other conditions and 48 (15.2%) did not have AECOPD after adjudication. Increased sputum purulence (OR 3.35, 95%CI 1.68-6.69) and elevated venous pCO2 (OR 1.04, 95%CI 1.01 - 1.07) were associated with a diagnosis of AECOPD but these were not associated with AECOPD alone without concomitant conditions. Radiographic evidence of pleural effusion (OR 0.26, 95%CI 0.12 - 0.58) was negatively associated with AECOPD with or without other conditions while radiographic evidence of pulmonary edema (OR 0.31; 95%CI 0.11 - 0.91) and lobar pneumonia (OR 0.13, 95%CI 0.07 - 0.25) suggested against the diagnosis of AECOPD alone.ConclusionThe study highlighted the complexity and difficulty of AECOPD diagnosis. A more specific clinical tool to diagnose AECOPD is needed.

Project description:The study developed accurate explainable machine learning (ML) models for predicting first-time acute exacerbation of chronic obstructive pulmonary disease (COPD, AECOPD) at an individual level. We conducted a retrospective case-control study. A total of 606 patients with COPD were screened for eligibility using registry data from the COPD Pay-for-Performance Program (COPD P4P program) database at Changhua Christian Hospital between January 2017 and December 2019. Recursive feature elimination technology was used to select the optimal subset of features for predicting the occurrence of AECOPD. We developed four ML models to predict first-time AECOPD, and the highest-performing model was applied. Finally, an explainable approach based on ML and the SHapley Additive exPlanations (SHAP) and a local explanation method were used to evaluate the risk of AECOPD and to generate individual explanations of the model's decisions. The gradient boosting machine (GBM) and support vector machine (SVM) models exhibited superior discrimination ability (area under curve [AUC] = 0.833 [95% confidence interval (CI) 0.745-0.921] and AUC = 0.836 [95% CI 0.757-0.915], respectively). The decision curve analysis indicated that the GBM model exhibited a higher net benefit in distinguishing patients at high risk for AECOPD when the threshold probability was <0.55. The COPD Assessment Test (CAT) and the symptom of wheezing were the two most important features and exhibited the highest SHAP values, followed by monocyte count and white blood cell (WBC) count, coughing, red blood cell (RBC) count, breathing rate, oral long-acting bronchodilator use, chronic pulmonary disease (CPD), systolic blood pressure (SBP), and others. Higher CAT score; monocyte, WBC, and RBC counts; BMI; diastolic blood pressure (DBP); neutrophil-to-lymphocyte ratio; and eosinophil and lymphocyte counts were associated with AECOPD. The presence of symptoms (wheezing, dyspnea, coughing), chronic disease (CPD, congestive heart failure [CHF], sleep disorders, and pneumonia), and use of COPD medications (triple-therapy long-acting bronchodilators, short-acting bronchodilators, oral long-acting bronchodilators, and antibiotics) were also positively associated with AECOPD. A high breathing rate, heart rate, or systolic blood pressure and methylxanthine use were negatively correlated with AECOPD. The ML model was able to accurately assess the risk of AECOPD. The ML model combined with SHAP and the local explanation method were able to provide interpretable and visual explanations of individualized risk predictions, which may assist clinical physicians in understanding the effects of key features in the model and the model's decision-making process.