Dataset Information

Computed tomography-based radiomics improves non-invasive diagnosis of Pneumocystis jirovecii pneumonia in non-HIV patients: a retrospective study.

ABSTRACT:

Background

Pneumocystis jirovecii pneumonia (PCP) could be fatal to patients without human immunodeficiency virus (HIV) infection. Current diagnostic methods are either invasive or inaccurate. We aimed to establish an accurate and non-invasive radiomics-based way to identify the risk of PCP infection in non-HIV patients with computed tomography (CT) manifestation of pneumonia.

Methods

This is a retrospective study including non-HIV patients hospitalized for suspected PCP from January 2010 to December 2022 in one hospital. The patients were randomized in a 7:3 ratio into training and validation cohorts. Computed tomography (CT)-based radiomics features were extracted automatically and used to construct a radiomics model. A diagnostic model with traditional clinical and CT features was also built. The area under the curve (AUC) were calculated and used to evaluate the diagnostic performance of the models. The combination of the radiomics features and serum β-D-glucan levels was also evaluated for PCP diagnosis.

Results

A total of 140 patients (PCP: N = 61, non-PCP: N = 79) were randomized into training (N = 97) and validation (N = 43) cohorts. The radiomics model consisting of nine radiomic features performed significantly better (AUC = 0.954; 95% CI: 0.898-1.000) than the traditional model consisting of serum β-D-glucan levels (AUC = 0.752; 95% CI: 0.597-0.908) in identifying PCP (P = 0.002). The combination of radiomics features and serum β-D-glucan levels showed an accuracy of 95.8% for identifying PCP infection (positive predictive value: 95.7%, negative predictive value: 95.8%).

Conclusions

Radiomics showed good diagnostic performance in differentiating PCP from other types of pneumonia in non-HIV patients. A combined diagnostic method including radiomics and serum β-D-glucan has the potential to provide an accurate and non-invasive way to identify the risk of PCP infection in non-HIV patients with CT manifestation of pneumonia.

Trial registration

ClinicalTrials.gov (NCT05701631).

SUBMITTER: Yu H

PROVIDER: S-EPMC10762815 | biostudies-literature | 2024 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Computed tomography-based radiomics improves non-invasive diagnosis of Pneumocystis jirovecii pneumonia in non-HIV patients: a retrospective study.

Yu Hang H Yang Zhen Z Wei Yuanhui Y Shi Wenjia W Zhu Minghui M Liu Lu L Wang Miaoyu M Wang Yueming Y Zhu Qiang Q Liang Zhixin Z Zhao Wei W Chen Liang-An LA

BMC pulmonary medicine 20240102 1

<h4>Background</h4>Pneumocystis jirovecii pneumonia (PCP) could be fatal to patients without human immunodeficiency virus (HIV) infection. Current diagnostic methods are either invasive or inaccurate. We aimed to establish an accurate and non-invasive radiomics-based way to identify the risk of PCP infection in non-HIV patients with computed tomography (CT) manifestation of pneumonia.<h4>Methods</h4>This is a retrospective study including non-HIV patients hospitalized for suspected PCP from Janu ...[more]

PMID: 38167022

Similar Datasets

Project description:Objectives: Pneumocystis jirovecii pneumonia (PCP) is an AIDS-defining illness. In patients with HIV, the benefit of PCP prophylaxis is well-defined when the CD4 T-cell count decreases below 200 cells/μL. In other immunocompromised patients, the value of PCP prophylaxis is not always as well-established. This study aimed to describe the epidemiology of PCP in recent years and assess how many patients with PCP did or did not receive prophylaxis in the month preceding the infection. Material and Methods: A multicenter retrospective study was performed in 3 tertiary care hospital. A list of patients that underwent broncho-alveolar lavage sampling and Pneumocystis jirovecii (PJ) PCR testing was retrieved from the microbiology laboratories. An in-house PJ quantitative PCR (qPCR) was used in each center. A cycle threshold (Ct) value of ≤ 28.5-30 was considered a probable PCP. For patients with a positive PJ qPCR but above this threshold, a predefined case definition of possible PCP was defined as a qPCR Ct value ≤ 34-35 and both of the following criteria: 1. Clinical and radiological features compatible with PCP and 2. The patient died or received PCP therapy and survived. Patient files from those with a qPCR Ct value ≤ 35 were reviewed to determine whether the patient fulfilled the case definition and if PCP prophylaxis had been used in the weeks preceding the PCP. Disease-specific guidelines, as well as hospital-wide guidelines, were used to evaluate if prophylaxis could be considered indicated. Results: From 2012 to 2018, 482 BAL samples were tested. Two hundred and four had a qPCR Ct value ≤ 35 and were further evaluated: 90 fulfilled the definition of probable and 63 of possible PCP while the remaining 51 were considered colonized. Seventy-four percentages of the patients with PCP were HIV-negative. Only 11 (7%) of the 153 patients had received prophylaxis, despite that in 133 (87%) cases prophylaxis was indicated according to guidelines. Conclusion: In regions where HIV testing and treatment is available without restrictions, PCP is mainly diagnosed in non-HIV immunocompromised patients. More than four out of five patients with PCP had not received prophylaxis. Strategies to improve awareness of antimicrobial prophylaxis guidelines in immunocompromised patients are urgently needed.

Project description:BACKGROUND:Thoracic ultrasound is helpful to evaluate lung pathology in patients with acute dyspnea. Several studies have demonstrated the efficacy of point-of-care ultrasound in patients with extrapulmonary TB and HIV co-infection. This retrospective, open-label case-control study explores the role of lung ultrasound in the diagnosis of Pneumocystis jirovecii pneumonia (PJP) in HIV-positive patients. In particular, it highlights the potential role of specific sonographic features that may be unique to this population. METHODS:The record of all HIV-positive patients admitted from 1.1.2013 to 31.6.2017 to the Department of Infectious Diseases and Tropical Medicine of san Bortolo Hospital, Vicenza, Italy, with a discharge diagnosis of acute lung injury (ALI) and who received point-of-care ultrasound of the chest for clinical purposes was included in the analysis. The patients were scanned according with the evidence-based recommendation. RESULTS:Of 273 HIV-positive patients whose records were reviewed, 81 (29.6%) were diagnosed with ALI. Complete documentation was available for 24 patients, of which 14 (58.3%) had microbiologically confirmed PJP (PJP+) and 10 (41.7%) had other conditions (PJP-). B-lines, subpleural consolidations, and cystic changes were significantly more frequent in patients with PJP (14/14 vs. 6/10, p = 0.0198; 14/14 vs. 4/10, p = 0.0016; 8/14 vs. 0/10, p = 0.0019, respectively). In particular, B-lines and subpleural consolidations were present in all PJP+ patients in our cohort giving a sensitivity of 100%, but their specificity was low (45 and 60%, respectively). On the contrary, the presence of consolidations with cystic changes had a very high specificity for PJP (100%), but low sensitivity (57%). Pleural effusions and consolidations with linear air bronchograms were not observed in PJP+ patients. CONCLUSIONS:B-lines, subpleural consolidations, and cystic changes are suggestive of PJP. Lung consolidation with air bronchograms and pleural effusion should prompt suspicion of other etiologies. These findings have the potential to be useful in the daily management of HIV-positive patients in resource-limited settings where other diagnostic tools are rarely available.

Dataset Information

Computed tomography-based radiomics improves non-invasive diagnosis of Pneumocystis jirovecii pneumonia in non-HIV patients: a retrospective study.

Background

Methods

Results

Conclusions

Trial registration

Publications

Computed tomography-based radiomics improves non-invasive diagnosis of Pneumocystis jirovecii pneumonia in non-HIV patients: a retrospective study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets