The predictive value of systemic inflammatory markers in 902 patients with tunneled hemodialysis catheter

Abstract Aim This study aimed to assess the predictive role of neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, lymphocyte-to-monocyte ratio, and mean platelet volume, on catheter survival in chronic hemodialysis patients, analyzing both infectious and non-infectious complications. Methods A retrospective analysis encompassed 1279 tunneled catheter insertion procedures involving 902 patients between March 2014 and October 2018. Patients were categorized into two main groups: (i) initial placement and (ii) exchange. The exchange group was further stratified into four subgroups: infection, dysfunction, displacement, and transitioning temporary hemodialysis catheters to long-term ones. Hematologic ratios were calculated from baseline hemogram data, including neutrophil, lymphocyte, monocyte, and platelet counts, while mean platelet volume was derived from the same hemogram. Results The patients in the exchange group displayed significantly higher lymphocyte and monocyte values ( p < 0.001), while lower values were noted for neutrophil–lymphocyte ratio and platelet-lymphocyte ratio ( p < 0.001). The transition group displayed higher monocyte values and lower mean platelet volume and lymphocyte-monocyte ratio values ( p < 0.05). In the infection-related exchange subgroup, higher neutrophil count, mean platelet volume, neutrophil–lymphocyte ratio, and platelet-lymphocyte ratio values were observed compared to other groups ( p < 0.05). Cases related to catheter dysfunction exhibited increased lymphocyte-monocyte ratio but lower neutrophil, monocyte, neutrophil–lymphocyte ratio, and platelet-lymphocyte ratio values ( p < 0.05). Conclusion This study highlights the interest of specific inflammatory markers, particularly monocytes, neutrophil–lymphocyte ratio, and platelet-lymphocyte ratio, in the management of tunneled catheters, notably in patients undergoing exchanges. However, cut-off values, essential for constructing management algorithms, are currently lacking, and prospective multicenter studies are needed for further elucidation. Graphical abstract