Growth Differentiation Factor 15 and Risk of Death in Haemodialysis Patients

Christelle Calen; Seraina Von Moos; Pietro Cippà; Alexandre Mebazaa; Mattia Arrigo

doi:10.1155/2023/5163548

Growth Differentiation Factor 15 and Risk of Death in Haemodialysis Patients

Int J Nephrol. 2023 Mar 23:2023:5163548. doi: 10.1155/2023/5163548. eCollection 2023.

Authors

Christelle Calen¹, Seraina Von Moos², Pietro Cippà³, Alexandre Mebazaa⁴, Mattia Arrigo⁵

Affiliations

¹ Division of General Internal Medicine, Hirslanden Zurich, Zurich, Switzerland.
² Department of Nephrology, University Hospital Zurich, Zurich, Switzerland.
³ Division of Nephrology, Ente Ospedaliero Cantonale, Ospedale Regionale di Lugano, Lugano, Switzerland.
⁴ Department of Critical Care and Anesthesiology, Lariboisiere University Hospital, and University of Paris, Paris, France.
⁵ Department of Internal Medicine, Stadtspital Zurich Triemli, Zurich, Switzerland.

Abstract

Aim: Noninvasive identification of haemodialysis patients at high risk of cardiovascular events and death might improve their outcome. Growth differentiation factor 15 is a prognostic biomarker in multiple disease entities, including cardiovascular disease. The aim of this study was to assess the association between plasma GDF-15 and mortality in a cohort of haemodialysis patients.

Methods: Circulating GDF-15 was measured in 30 patients after a regular haemodialysis session, followed by a clinical follow-up for all-cause death. Measurements were performed using the Proseek Multiplex Cardiovascular disease panels (Olink Proteomics AB) and validated using the Elecsys GDF-15 electrochemiluminescence immunoassay on a Cobas E801 analyzer (Roche Diagnostics).

Results: During a median of 38 months, 9 patients (30%) died. Seven deaths occurred in the group of patients with a circulating GDF-15 above the median and two in the group with lower GDF-15. Mortality was significantly higher in patients with circulating GDF-15 levels above the median, log-rankP = 0.044. The performance of circulating GDF-15 to predict long-term mortality has an area under the ROC curve of 0.76, P = 0.028. Prevalence of most relevant comorbidities and the Charlson comorbidity index were similar across the two groups. A high agreement with a correlation among both diagnostic methods was observed (Spearman's rho = 0.83, P < 0.001).

Conclusion: Plasma GDF-15 displays promising prognostic properties for the prediction of long-term survival beyond clinical parameters in patients on maintenance haemodialysis.