Copeptin predicts 10-year all-cause mortality in community patients: a 10-year prospective cohort study

Odermatt, Jonas ; Bolliger, Rebekka ; Hersberger, Lara ; Ottiger, Manuel ; Christ-Crain, Mirjam ; Briel, Matthias ; Bucher, Heiner C. ; Mueller, Beat ; Schuetz, Philipp

In: Clinical Chemistry and Laboratory Medicine (CCLM), 2016, vol. 54, no. 10, p. 1681-1690

Add to personal list
    Summary
    AbstractBackground: Copeptin, the C-terminal part of the arginine vasopressin (AVP) precursor peptide, is secreted in response to stress and correlates with adverse clinical outcomes in the acute-care hospital setting. There are no comprehensive data regarding its prognostic value in the community. We evaluated associations of copeptin levels with 10-year mortality in patients visiting their general practitioner (GP) for a respiratory infection included in a previous trial. Methods: This is a post hoc analysis including data from 359 patients included in the PARTI trial. Copeptin was measured in batch-analysis on admission and after 7 days. We calculated Cox regression models and area under the receiver operating characteristic curve (AUC) to assess an association of copeptin with mortality and adverse outcome. Follow-up data were collected by GP, patient and relative tracing through phone interviews 10 years after trial inclusion. Results: After a median follow-up of 10.0 years, mortality was 9.8%. Median admission copeptin levels (pmol/L) were significantly elevated in non-survivors compared to survivors (13.8, IQR 5.9-27.8; vs. 6.3 IQR 4.1-11.5; p<0.001). Admission copeptin levels were associated with 10-year all-cause mortality [age-adjusted hazard ratio 1.7 (95% CI, 1.2-2.5); p<0.001, AUC 0.68]. Results were similar for discharge copeptin levels. Copeptin also predicted adverse outcomes defined as death, pulmonary embolism and major adverse cardiac and cerebrovascular events. Conclusions: In a sample of community-dwelling patients visiting their GP for a respiratory infection, copeptin levels were associated with 10-year all-cause mortality. In conjunction with traditional risk factors, this marker may help to better direct preventive measures in this population.