The Guiding Value Of Procalcitonin In Diagnosis And Treatment Of Ventilator Associated Pneumonia

Objective To assess the guiding value of procalcitonin (PCT) in diagnosis and treatment of ventilator associated pneumonia(VAP).Methods In a randomized intervention trial, sixty patients on ventilators were divided into VAP group (n=30) and non-VAP group (n=30) depending on whether the patients developed VAP in 7 days or not. All the patients were evaluated with acute physiology and chronic health evaluationⅡscores (APACHEⅡ) at admission to ICU, and clinical pulmonary infection score (CPIS) was calculated when they developed VAP. The VAP group again was divided into PCT group (n=15)and the control group (n=15).Data of PCT, C-reactive protein (CRP) and WBC count were assessed at baseline, in 6 hours when VAP was suspected, and every two days after mechanical ventilation. Chest roentgenogram (CR), sputum culture and hemoculture were taken timely. The PCT levels were graded into four categories, i.e.< 0.5u g/l,0.5-2.0 ug/l, 2.0-10.0 ug/l and≥10.0 ug/l. The control group received antibiotics according to usual practice and stopped when CPIS≤6 scores. In the PCT group, antibiotic treatment was based on serum PCT concentrations as follows:strongly discouraged, less than 0.1 ug/l; discouraged, less than 0.25 ug/l; encouraged, greater than 0.25 ug/l; strongly encouraged, greater than 0.5ug/l.Results The CRP and WBC count showed no obvious difference between the VAP and non-VAP group before mechanical ventilation (all P> 0.05), after ventilation both mean values increased, with the P 0.046 and 0.822, respectively; Taking CRP≥28mg/l and WBC count≥10×109/l as the cutoff value, the diagnostic sensitivity of CRP and WBC count for VAP were 73.3%(22/30 cases) and 66.7%(20/30 cases) respectively, their specificity was 50% (15/30 cases) and 43.3%(13/30 cases), respectively. Their positive predictive values were 59.4%(22/37cases) and 54.1%(20/37cases), respectively; and their negative predictive values were 65.2%(15/23 cases) and 56.5%(13/23 cases),respectively.Taking PCT≥0.40ug/l as the cutoff value, the PCT positive percentage did not show difference between VAP group and non-VAP group before mechanical ventilation (P> 0.05). However, it was much higher in the VAP group than that of non-VAP group after mechanical ventilation (X2=27.778, P<0.001). A PCT level of 0.40 ug/1 yielded a 77.8%(28/36 cases) positive predictive value and a 91.7%(22/24cases) negative predictive value for VAP in the study cohort (AUROCC=0.823; 95%CI,0.71-0.94; p<0.001),with the sensitivity of 93.3%(28/30 cases) and the specificity of 73.3%(22/30cases). The antibiotic duration of the PCT group was 12.6+/-5.6 days compared with 15.1+/-9.1 days for the control group(p< 0.05). Patients assigned to the PCT group had 2.5 days shorter mean duration of antibiotic therapy for the first episode of infection than the control group (p<0.05). The PCT level on Day 1 and the APACHEⅡwere significantly higher in patients who died in 28 days (all P<0.05). Moreover, PCT levels retained or increased in those who had persistent infection or died.Conclusion PCT has high sensitivity and specificity in the diagnosis of VAP. A CPIS≥6 combined with serum levels of procalcitonin≥0.40ug/l markedly improved the specificity (100%). A high APACHEⅡand a retained or increased PCT level predicted unfavorable outcome. PCT guidance substantially reduces antibiotic use in VAP, so that timely surveillance of serum PCT is necessary for patients on ventilator.