(179) Inflammation based scoring systems and ratios to identify cystic fibrosis patients at risk for intravenous antibiotic therapy

Background: Pulmonary disease in cystic fibrosis (CF) is characterized by chronic bacterial infection with recurrent pulmonary exacerbations (PEx), which are associated with an elevation of systemic inflammatory markers. As PEx are associated with higher morbidity and mortality, their prevention is an important therapeutic goal in the treatment of CF patients. Identifying patients at risk for future PEx can be challenging in clinical practice. The objective of this study was therefore to compare established and novel markers of chronic inflammation and their ability to predict the need for intravenous antibiotic therapy (nIVT), as surrogate marker for PEx or decline in pulmonary status in patients with CF.

Methods: In this retrospective, longitudinal study patients ≥18 years treated between 2014 - 2018 in the adult CF Center of the university hospital Munich, Germany, who had at least two pulmonary function tests per year for a 3 year period, were included. Baseline was defined as the first year after the first measurement. The follow-up period was the time between baseline and the end of follow-up (FU). Patients with and without nIVT during FU were compared using the following scoring systems or ratios: Neutrophil/Lymphocyte (NLR), Lymphocyte/Monocyte (LMR), CRP/albumin, Glasgow prognostic score (GPS) and CF-Able score. In the univariate analysis, we used chi-square and fisher exact test to compare categorical scores and t-test to compare numerical scores. In the multivariate analysis, we used logistic regression models, adjusted for age, sex, nIVT at baseline, BMI, homozygote delF508 mutation, diabetes mellitus, chronic pulmonary infection (Pseudomonas aeruginosa, Stenotrophomonas maltophilia, Mycobacterium abscessus, Achromobacter xylosoxidans) and use of a CFTR modulator therapy. Models AICs (Akaike information criterion) were compared to identify the marker with the best prediction ability. Results are reported as Odds-ratios (OR) with p<span lang="EN-US" style="font-size: 11.0pt; line-height: 107%; font-family: 'Calibri',sans-serif; mso-ascii-theme-font: minor-latin; mso-fareast-font-family: Calibri; mso-fareast-theme-font: minor-latin; mso-hansi-theme-font: minor-latin; mso-bidi-font-family: 'Times New Roman'; mso-bidi-theme-font: minor-bidi; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;">-values.

Results: We evaluated 131 patients with CF. 43.5% were female, the mean age was 34.0 years. 75 patients (57.3%) had at least one nIVT during FU. In the univariate analysis, having a nIVT during FU was significantly associated with higher scores for inflammation in all markers examined. In the multivariate analysis the following markers at baseline were significantly associated with having a nIVT during FU: CRP (OR = 7.29, p = 0.01), CRP/albumin (OR = 1.09, p = 0.004), decreased LMR% (OR = 0.55, p = 0.03), increased NLR% (OR = 1.51, p = 0.01), GPS of 1 vs. 0 (OR = 3.52, p = 0.02), and a CF-Able score of 2-4.5 vs. < 2 (OR=3.82, p = 0.02). Comparison of the different regression models showed that the CRP/albumin ratio was superior to the other models examined in differentiating patients with or without nIVT in the FU period.

Conclusions: Our retrospective study shows that several inflammation-based scoring systems or ratios established for other disease entities are able to identify patients with nIVT in CF patients. Although the CF-Able score was not designed to predict PEx, it does include some variables associated with increased risk for PEx and, correspondingly, nIVT. Interestingly, the easily calculated CRP/albumin ratio is superior to the CF-Able Score in distinguishing CF patients with or without nIVT and thus represents an interesting possibility for clinical use.

Acknowledgements:

References: