Bronchoalveolar lavage cytokines are of minor value to diagnose complications following lung transplantation

Abstract

Early diagnosis and treatment of acute cellular rejection (ACR) may improve long-term outcome for lung transplant recipients (LTRs). Cytokines have become valuable diagnostic tools in many medical fields. The role of bronchoalveolar lavage (BAL) cytokines is of unknown value to diagnose ACR and distinguish rejection from infection. We hypothesized that distinct cytokine patterns obtained by surveillance bronchoscopies during the first year after transplantation are associated with ACR and microbiologic findings. We retrospectively analyzed data from 319 patients undergoing lung transplantation at University Hospital Zurich from 1998 to 2016. We compared levels of IL-6, IL-8, IFN-γ and TNF-α in 747 BAL samples with transbronchial biopsies (TBB) and microbiologic results from surveillance bronchoscopies. We aimed to define reference values that would allow distinction between four specific groups "ACR", "infection", "combined ACR and infection" and "no pathologic process". No definitive pattern was identified. Given the overlap between groups, these four cytokines are not suitable diagnostic markers for ACR or infection after lung transplantation.

Keywords: Acute cellular rejection; Bronchoalveolar lavage; Cytokine; Diagnosis; Lung transplantation; Standardization.

Fig. 1

Concentration of IL-6 in BAL fluid for different specific groups. The log transformed concentrations (log([pg/ml]) of IL-6 at the time of ≥A1 rejection; viral, bacterial or fungal infection; “no pathologic process” or “combined ACR and infection” is shown as box-and-whiskers plots. Significance for differences between specific groups was determined using Kruskal-Wallis test. Significance for differences between two categories was determined using Wilcoxon Rank Sum test. Log levels of IL-6 were lowest during ACR and highest during “combined ACR and infection” compared with “no pathologic process” (p = 0.87 and p = 0.11, respectively).

Fig. 2

Left: ROC curves for IFN-γ, IL-6, IL-8 and TNF-α to diagnose isolated ACR in surveillance bronchoscopies with TBB during the first year after transplantation. Area under the curve was 0.48 (IFN-γ), 0.55 (IL-6), 0.51 (IL-8) and 0.52 (TNF-α). No optimal cut-off can be derived from these data. Right: ROC curves for IL-6, IL-8, IFN-γ and TNF-α to diagnose “combined ACR and infection” in surveillance bronchoscopies with TBB during the first year after transplantation. Area under the curve for IFN-γ was 0.517 and for IL-8 was 0.531. Area under the curve for TNF-α was 0.509. No optimal cutoff can be derived from these data, particularly where higher scores of TNF-α do not appear to correspond with higher probability of “combined ACR and infection”.

Fig. 3

Kaplan-Meier analysis of survival after lung transplantation. No significant differences in overall survival were observed between patients with and without at least one episode ACR diagnosed in surveillance bronchoscopies during the first year after transplantation (p = 0.54).