Clinical Features and Prediction Model of Secondary Infection Risk in Adult Patients With Chronic Respiratory Diseases: A Case-Control Study

He Qing Huang, Hong Lü, Yi‐Ling Chen, M. Shen, Zengqiang Chen, Jie XuJournal of Clinical Medicine Research 2025 · DOI

Abstract

This study focused and explored the common features between secondary infections of various chronic respiratory diseases. The prediction model is expected to enable timely detection and treatment of secondary infections in clinical practice.

Introduction

Chronic respiratory diseases (CRDs)—including chronic obstructive pulmonary disease (COPD), interstitial pneumonia, and bronchiectasis—are the third leading cause of death worldwide. These conditions often result in permanent damage to lung architecture and weakened local immunity, making patients highly susceptible to secondary bacterial and fungal infections.

A significant clinical challenge lies in the timely diagnosis of these infections, as symptoms such as cough, wheezing, and chest tightness often mirror the exacerbation of the underlying chronic disease. While many studies focus on specific conditions like COPD, there is a lack of research on general predictive markers that apply across the broader spectrum of CRDs. This paper presents a retrospective case-control study aimed at identifying common clinical "red flags" and developing a multivariate prediction model to help clinicians identify high-risk patients earlier.

Methods

The authors conducted a retrospective case-control study at the First Affiliated Hospital of Soochow University, enrolling adult inpatients between 2019 and 2022.

Flowchart of patient selection and study grouping.

As shown in the selection process, the final cohort of 154 patients was divided into two groups:

  • Observation Group (n=54): Patients with confirmed secondary infection via pathogen culture.
  • Control Group (n=100): Patients with no infection or normal flora.

The research team analyzed demographic data, comorbidities, clinical symptoms, laboratory indicators (such as WBC, CRP, and PCT), and echocardiography results. Multivariate logistic regression was used to identify independent risk factors, and predictive performance was evaluated using Receiver Operating Characteristic (ROC) curves.

Key Insights and Results

The study identified three primary independent risk factors that, when combined, formed a robust predictive model:

  • Nutritional Status (BMI): Lower Body Mass Index (malnutrition) was strongly correlated with infection (OR = 0.789).
  • Immune Status (Lymphocyte Count): Decreased lymphocyte counts (immunosuppression) were a major risk factor (OR = 0.122).
  • Cardiovascular/Pulmonary Pressure (TRPG): An increased Tricuspid Regurgitation Pressure Gradient (TRPG) was a significant predictor (OR = 1.116). This suggests that as chronic lung disease leads to higher pulmonary vascular resistance and right-heart pressure, the risk of secondary infection increases.

Pathogen Profile

Pseudomonas aeruginosa was the most common bacterial pathogen (found in 37% of infected cases), particularly in COPD patients. Aspergillus fumigatus was identified as the primary cause of filamentous fungal infections.

Model Performance

The combined prediction model (BMI + Lymphocytes + TRPG) achieved an AUC of 0.788 (95% CI: 0.689–0.887, P = 0.000), outperforming any single indicator. Notably, C-reactive protein (CRP) proved to be a more sensitive indicator of early infection than Procalcitonin (PCT) in this population, possibly because PCT levels can be masked by premature antibiotic use or take longer to peak.

Discussion and Future Work

This study provides a non-invasive method to identify high-risk patients who may require earlier antibiotic intervention, potentially reducing the medical and economic burden of disease exacerbation. However, the authors acknowledge limitations including the retrospective design and a relatively small observation group.

Future work should involve multi-center prospective cohort studies to validate the model. The authors also suggest incorporating additional physiological markers, such as hemoglobin levels and more granular respiratory function tests, to further refine the tool's accuracy.

Summary updated

Abstract

This study focused and explored the common features between secondary infections of various chronic respiratory diseases. The prediction model is expected to enable timely detection and treatment of secondary infections in clinical practice.
BibTeX
@article{huang2025clinical,
  author = {He Qing Huang and Hong Lü and Yi‐Ling Chen and M. Shen and Zengqiang Chen and Jie Xu},
  title = {{Clinical Features and Prediction Model of Secondary Infection Risk in Adult Patients With Chronic Respiratory Diseases: A Case-Control Study}},
  journal = {Journal of Clinical Medicine Research},
  year = {2025},
  doi = {10.14740/jocmr6291},
}
Figure