# Current Blood Eosinophilia Does Not Predict the Presence of Pulmonary Hypertension in Patients with End-Stage Lung Disease

**Authors:** Michaela Barnikel, Nikolaus Kneidinger, Michael Gerckens, Carlo Mümmler, Alexandra Lenoir, Pontus Mertsch, Tobias Veit, Gabriela Leuschner, Andrea Waelde, Claus Neurohr, Jürgen Behr, Katrin Milger

PMC · DOI: 10.3390/jcm14041120 · 2025-02-09

## TL;DR

This study found that high blood eosinophil levels do not reliably predict pulmonary hypertension in patients with advanced lung diseases like COPD, CF, and ILD.

## Contribution

The study is the first to show that current blood eosinophilia is not a useful predictor of pulmonary hypertension in end-stage lung disease patients.

## Key findings

- Severe pulmonary hypertension was more common in ILD patients compared to COPD and CF.
- Eosinophil levels were not significantly correlated with hemodynamic parameters in any of the studied lung diseases.
- Eosinophilic COPD was associated with unclassified PH, higher cardiac output, and greater lung volumes.

## Abstract

Objectives: To investigate the role of blood eosinophils in predicting PH in end-stage lung disease. Methods: We conducted a retrospective study of adults with CF, COPD, and ILD who underwent RHC during lung transplant evaluations (2010–2022). Patients were classified by the 2022 ECS/ERS PH guidelines with pulmonary function and laboratory tests, including hemograms. The eosinophil threshold was set at 0.30 G/L. Results: We analyzed 663 patients (n = 89 CF, n = 294 COPD, and n = 280 ILD). Severe PH was more common in ILD (16%) than in CF (4%) and COPD (7%) (p = 0.0002), with higher eosinophil levels in ILD (p = 0.0002). No significant correlation was found between eosinophil levels and hemodynamic parameters (PAPm, PVR, and CI) across CF, COPD, and ILD (PAPm: p = 0.3974, p = 0.4400 and p = 0.2757, respectively; PVR: p = 0.6966, p = 0.1489 and p = 0.1630, respectively; CI: p = 0.9474, p = 0.5705 and p = 0.5945, respectively), nor was a correlation observed in patients not receiving OCS. Linear regression analysis confirmed the lack of association (PAPm: p = 0.3355, p = 0.8552 and p = 0.4146, respectively; PVR: p = 0.6924, p = 0.8935 and p = 0.5459, respectively; CI: p = 0.4260, p = 0.9289 and p = 0.5364, respectively), controlling for 6-MWD, Nt-proBNP, and ICS/OCS dosages. ROC analysis indicated eosinophils were ineffective in distinguishing PH severity levels across these diseases (AUC 0.54, 0.51, and 0.53, respectively). The analysis of eosinophil levels measured 18 ± 6 months prior to baseline found no predictive correlation with the presence of PH either. Eosinophil levels did not differ significantly among PH groups, but eosinophilic COPD was linked to more unclassified PH, higher CO, and greater lung volumes than non-eosinophilic COPD. Conclusions: In our cohort of end-stage CF, COPD, and ILD patients, blood eosinophilia did not predict the presence of PH but was associated with hemodynamic parameters and lung volumes in COPD.

## Linked entities

- **Diseases:** CF (MONDO:0009061), COPD (MONDO:0005002), ILD (MONDO:0015925), pulmonary hypertension (MONDO:0005149)

## Full-text entities

- **Genes:** PVR (PVR cell adhesion molecule) [NCBI Gene 5817] {aka CD155, HVED, NECL5, Necl-5, PVS, TAGE4}
- **Diseases:** ILD (MESH:D017563), end-stage CF (MESH:D007676), COPD (MESH:D029424), Blood Eosinophilia (MESH:D004802), CF (MESH:D003550), Pulmonary Hypertension (MESH:D006976), End-Stage Lung Disease (MESH:D058625)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Figures

3 figures with captions in the complete paper: https://tomesphere.com/paper/PMC11856528/full.md

---
Source: https://tomesphere.com/paper/PMC11856528