Factors associated with severe disease and in-hospital mortality in children with non-HIV-associated Pneumocystis jirovecii pneumonia: a single-center retrospective cohort study
4 affiliations
- 1Department of Scientific Research, Zigong First People's Hospital, Zigong, Sichuan, China
- 2Department of Pediatrics, West China Second University Hospital, Sichuan University, Chengdu, Sichuan, China
- 3Department of Pediatrics, Meishan People’s Hospital, Meishan, Sichuan, China
- 4Department of Pediatrics, Zigong First People’s Hospital, Zigong, Sichuan, China
Abstract
Objective: To identify factors associated with severe Pneumocystis jirovecii pneumonia (PJP) and in-hospital mortality and evaluate their ability to discriminate severe PJP in children without human immunodeficiency virus (HIV) infection.
Methods: This single-center retrospective cohort study included 61 children with non-HIV-associated PJP admitted between February 2022 and June 2025, including 31 with non-severe and 30 with severe PJP. Univariable and multivariable Firth penalized logistic regression identified factors associated with severe PJP, and receiver operating characteristic curves assessed discrimination. A restricted Firth logistic model explored in-hospital mortality among severe cases.
Results: Thirty children (49.2%) had severe PJP, and nine (14.8%) died during hospitalization; all deaths occurred in the severe group. Compared with the non-severe group, the severe group had more frequent primary immunodeficiency, fever, dyspnea, and cytomegalovirus (CMV) coinfection; higher D-dimer, lactate dehydrogenase (LDH), procalcitonin, and CT severity scores; and lower albumin, PaO2/FiO2, and CD4+/CD8+ T-cell ratios (all P<0.05). In the primary multivariable model, LDH (per 100-U/L increase: adjusted odds ratio [aOR]=1.532, 95% confidence interval [CI]: 1.061–2.374; P = 0.021) and CMV coinfection (aOR=5.164, 95% CI: 1.099–28.048; P = 0.037) were associated with severe PJP. LDH remained significant in the sensitivity analysis, whereas CMV coinfection did not. LDH discriminated severe from non-severe PJP with an area under the curve of 0.868 (95% CI: 0.776–0.960); the optimal cutoff was 423.0 U/L, with 90.0% sensitivity and 71.0% specificity. The combined model had an AUC of 0.878 (95% CI: 0.791–0.966), without significant improvement over LDH alone (P = 0.701). Among severe cases, non-survivors had more frequent malnutrition and CMV coinfection, higher LDH and procalcitonin, and lower neutrophil counts, albumin, PaO2/FiO2, and CD4+/CD8+ T-cell ratios (all P<0.05). In the exploratory mortality model, LDH (per 100-U/L increase: aOR=1.654, 95% CI: 1.080–4.271; P = 0.018) and albumin (per 1-g/L increase: aOR=0.605, 95% CI: 0.251–0.878; P = 0.004) were associated with in-hospital mortality.
Conclusions: Elevated LDH was consistently associated with severe PJP and showed good discrimination. CMV coinfection was associated with severe PJP only in the primary model and provided no significant incremental discrimination. Elevated LDH and decreased albumin may be associated with in-hospital mortality. Larger external cohorts are needed to validate these findings.
Related
- Pneumocystis Jirovecii Pneumonia Associated with Prolonged CD4-positive T-cell Depletion after Bendamustine-Rituximab Therapy: A Case Report
- Association of PCR cycle threshold values in respiratory specimens with mechanical ventilation and mortality in HIV-negative pneumocystis jirovecii pneumonia
- Added Value of Next-Generation Sequencing for Multilocus Sequence Typing Analysis of a Pneumocystis jirovecii Pneumonia Outbreak1
- Very Late-Onset Cytomegalovirus Disease With Pneumocystis jirovecii Pneumonia Coinfection in a Renal Transplant Recipient: A Case Report and Narrative Review of Published Cases.
- Pneumocystis jirovecii Pneumonia in an Adolescent With Acute Lymphoblastic Leukemia Due to Non-adherence to Prophylaxis: A Case Report.
BibTeX
@article{ning2026factors,
title = {{Factors associated with severe disease and in-hospital mortality in children with non-HIV-associated Pneumocystis jirovecii pneumonia: a single-center retrospective cohort study}},
author = {Junjie Ning and Dan Chen and Zhidong Yu and Fajuan Tang and Lina Qiao},
journal = {Frontiers in Cellular and Infection Microbiology},
year = {2026},
month = oct,
doi = {10.3389/fcimb.2026.1937817},
url = {https://doi.org/10.3389/fcimb.2026.1937817},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Factors associated with severe disease and in-hospital mortality in children with non-HIV-associated Pneumocystis jirovecii pneumonia: a single-center retrospective cohort study
AU - Junjie Ning
AU - Dan Chen
AU - Zhidong Yu
AU - Fajuan Tang
AU - Lina Qiao
PY - 2026
DA - 2026/10/12
JO - Frontiers in Cellular and Infection Microbiology
DO - 10.3389/fcimb.2026.1937817
UR - https://doi.org/10.3389/fcimb.2026.1937817
AB - Objective: To identify factors associated with severe Pneumocystis jirovecii pneumonia (PJP) and in-hospital mortality and evaluate their ability to discriminate severe PJP in children without human immunodeficiency virus (HIV) infection. Methods: This single-center retrospective cohort study included 61 children with non-HIV-associated PJP admitted between February 2022 and June 2025, including 31 with non-severe and 30 with severe PJP. Univariable and multivariable Firth penalized logistic regression identified factors associated with severe PJP, and receiver operating characteristic curves assessed discrimination. A restricted Firth logistic model explored in-hospital mortality among severe cases. Results: Thirty children (49.2%) had severe PJP, and nine (14.8%) died during hospitalization; all deaths occurred in the severe group. Compared with the non-severe group, the severe group had more frequent primary immunodeficiency, fever, dyspnea, and cytomegalovirus (CMV) coinfection; higher D-dimer, lactate dehydrogenase (LDH), procalcitonin, and CT severity scores; and lower albumin, PaO2/FiO2, and CD4+/CD8+ T-cell ratios (all P<0.05). In the primary multivariable model, LDH (per 100-U/L increase: adjusted odds ratio [aOR]=1.532, 95% confidence interval [CI]: 1.061–2.374; P = 0.021) and CMV coinfection (aOR=5.164, 95% CI: 1.099–28.048; P = 0.037) were associated with severe PJP. LDH remained significant in the sensitivity analysis, whereas CMV coinfection did not. LDH discriminated severe from non-severe PJP with an area under the curve of 0.868 (95% CI: 0.776–0.960); the optimal cutoff was 423.0 U/L, with 90.0% sensitivity and 71.0% specificity. The combined model had an AUC of 0.878 (95% CI: 0.791–0.966), without significant improvement over LDH alone (P = 0.701). Among severe cases, non-survivors had more frequent malnutrition and CMV coinfection, higher LDH and procalcitonin, and lower neutrophil counts, albumin, PaO2/FiO2, and CD4+/CD8+ T-cell ratios (all P<0.05). In the exploratory mortality model, LDH (per 100-U/L increase: aOR=1.654, 95% CI: 1.080–4.271; P = 0.018) and albumin (per 1-g/L increase: aOR=0.605, 95% CI: 0.251–0.878; P = 0.004) were associated with in-hospital mortality. Conclusions: Elevated LDH was consistently associated with severe PJP and showed good discrimination. CMV coinfection was associated with severe PJP only in the primary model and provided no significant incremental discrimination. Elevated LDH and decreased albumin may be associated with in-hospital mortality. Larger external cohorts are needed to validate these findings.
SN - 2235-2988
ER -