Prediction about severity and outcome of sepsis by pro-atrial natriuretic peptide and pro-adrenomedullin / 中华创伤杂志(英文版)

ABSTRACT

<p><b>OBJECTIVE</b>Measurement of biomarkers is a potential approach to early prediction of the risk of mortality in patients with sepsis. The aim of the present study was to evaluate the prognostic value of pro-atrial natriuretic peptide (pro-ANP) and pro-adrenomedullin (pro-ADM) levels in a cohort of medical intensive care patients and to compare it with that of other known biomarkers and physiological scores.</p><p><b>METHODS</b>Blood samples of 51 consecutive critically ill patients admitted to the intensive care unit and 53 age-matched healthy control people were evaluated in this prospective study. The prognostic value of pro-ANP and pro-ADM levels was compared with that of acute physiology and chronic health evaluation (APACHE) II scores and various biomarkers such as C-reactive protein, interleukin-6 and procalcitonin. Pro-ANP and pro-ADM were detected by a new sandwich immunoassay.</p><p><b>RESULTS</b>On admission, 25 patients had systemic inflammatory response syndrome (SIRS), 12 sepsis, 9 severe sepsis and 5 septic shock. At that time, the median levels (ng/ml) of pro-ANP and pro-ADM were 87.22 and 0.34 respectively in patients with SIRS, 1533.30 and 2.23 in those with sepsis, 1098.73 and 4.57 in those with severe sepsis, and 1933.94 and 8.21 in those with septic shock. With the increasing severity of disease, the levels of pro-ANP and pro-ADM were gradually increased. On admission, the circulating levels of pro-ANP and pro-ADM in patients with sepsis, severe sepsis, or septic shock were significantly higher in non-survivors than in survivors (P less than 0.05). In a receiver operating characteristic curve analysis for the survival of patients with sepsis, the areas under the curve (AUCs) for pro-ANP and pro-ADM were 0.89 and 0.87 respectively, which was similar to the AUCs for procalcitonin and APACHE II scores.</p><p><b>CONCLUSION</b>Pro-ANP and pro-ADM are valuable biomarkers for prediction of severity of septic patients.</p>