RESUMEN
<p><b>OBJECTIVE</b>To assess the clinical value ofprocalcitonin in cirrhotic patients with severe infection by comparing the serum procalcitonin levels in those patients with and without liver cirrhosis when suffering from sepsis.</p><p><b>METHODS</b>A total of 225 septic patients were included in the study,including 91 patients without hepatopathy, 80 patients with cirrhosis, and 54 patients with chronic liver disease. The serum procalcitonin level was measured in all patients and statistically assessed for correlation with relevant clinical biochemistry indicators. The t-test, ANOVA test, Mann-Whitney U test, chi-square test and Spearman's correlation analysis were used for statistical analyses.</p><p><b>RESULTS</b>The patients with cirrhosis showed significantly lower serum procalcitonin levels (0.84 (0.32-3.44) ng/ml) than the patients with no hepatopathy (2.17 (0.70-9.18) ng/ml) or the patients with chronic liver disease (2.12 (0.33-13.61) ng/ml) (both P less than 0.05); the patients in the no hepatopathy group and the chronic liver disease group showed statistically similar levels of serum procalcitonin (P=0.616). The patients with cirrhosis of Child-Pugh grade C showed significantly higher level of serum procalcitonin (1.25 (0.54-4.61) ng/ml) than those patients with Child-Pugh grade B (0.33 (0.14-1.31) ng/ml; P=0.026), suggesting that patients with Child-Pugh C stage cirrhosis may be more susceptible to gram-negative bacterial infection. In the cirrhosis group,serum procalcitonin level was positively correlated with white blood cell (WBC) count (r=0.312) and percentage of neutrophils (N%) (r=0.228) (both P less than 0.05). Correlation analysis of the no hepatopathy group and the chronic liver disease group showed no correlation between serum procalcitonin level and either WBC or N%.</p><p><b>CONCLUSION</b>Under the sepsis condition, cirrhotic patients have lower serum procalcitonin level than patients without cirrhosis, and the serum procalcitonin level is positively correlated with WBC count and N%.</p>