| Object:The prognostic rules widely used in western countries such as pneumonia severity index(PSI),CURB-65 and sepsis score are helpful for the management of patients with community-acquired pneumoniae(CAP).However in the developing countries,there are rare data about the implement of these rules.The present study is performed to retrospectively evaluate the role of PSI,CURB-65,and sepsis score in the management of hospitalized CAP patients in a university affiliated hospital in China.Methods:A total of 675 CAP patients consecutively hospitalized in an affiliated hospital of Chinese university were retrospectively investigated.The patients were classified into two or three risk groups according to the definitions of PSI,CURB-65, and sepsis score.The ability of the three rules to predict mortality was compared by Receiver Operating Characteristic(ROC) curve while the risk factors of mortality and hospital length of stay(LOS) were studied.The effect of ICU treatment on outcome of severe CAP patients was also evaluated.Results:All of the three score systems showed that the mortality of the low-risk or moderate-risk group was significantly lower than high-risk group:0.7%vs 31.8%for PSI,2.5%vs 14.6%vs 60.5%for CURB-65,and 0.4%vs 4.8%vs 56.2%(all p < 0.01). The similar results were also found in ICU admission rate and LOS in hospital.The areas under the ROC curve in the prediction of mortality were 0.94,0.91,and 0.89 for PSI,CURB-65 and sepsis score respectively.Compared with the corresponding control groups,the mortality remarkably increased in patients with smoking,admission to ICU, respiratory failure or combined with heart diseases(all p < 0.05).The difference was also significant in LOS between control groups and patients with ICU treatment, respiratory failure,combined with heart or renal diseases(all p < 0.05).Furthermore, analysis of PSI showed that in the high-risk group,patients receiving ICU treatment right at admission were far severer than those who received non-ICU treatment or received ICU treatment after the failure of non-ICU treatment:the PSI scores were 186.0±34.7 vs 118.6±23.7 vs 145.8±30.7,respectively(both p < 0.05).All the three severity score systems showed that in the high-risk group,the mortality of patients who took ICU treatment right at admission was higher than non-ICU treatment,but much lower than patients who took ICU treatment after the failure of non-ICU treatment(all/? <0.05).Conclusions:All the three severity score systems have a high predictive value for mortality and LOS in hospitalised CAP patients.PSI seems to take a little more advantage in predicting mortality than CURB-65 and sepsis score.In regard to easy handle and instant decisions to determine the candidates for inpatient treatment, CURB-65 is the favorable prognostic score.We recommend that severe CAP patients take ICU treatment right at admission. |