ABSTRACT
Objective To examine the short-term prognostic value of procalcitonin ( PCT ) combined with coagulation factors for cirrhotic patients complicated with spontaneous bacterial peritonitis (SBP).Methods Clinical data of 128 cirrhotic patients complicated with SBP admitted in Jinhua Central Hospital from June 2014 to October 2017 were retrospectively analyzed .In 3 months after admission , 83 patients survived ( survival group ) and 45 patients died ( fatal group ) .The factors related to prognosis were analyzed with Logistic regression and the prediction model was constructed with the weights derived from regression coefficients.The ROC curve and the area under the curve (AUC) of combination of PCT with coagulation factors were used to predict the survival of patients .Results Univariate analysis indicated that the level of PCT , total bilirubin ( TBil ) , serum creatinine ( Scr ) , prothrombin time ( PT ) , prothrombin activity ( PTA ) , blood coagulation factor Ⅱ, Ⅴ, Ⅶ, Ⅸ, Ⅹ, Ⅺ and Ⅻ were factors affecting the prognosis of cirrhotic patients complicated with SBP (P<0.01).Multivariate analysis showed that PCT , blood coagulation factors Ⅴ and Ⅸ were independent factors of short-term prognosis of cirrhotic patients complicated with SBP.The constructed predictive model was Logit (P) =1.200+0.099 ×PCT-0.026 × clotting factor Ⅴ-0.038 ×clotting factor Ⅸ.The sensitivity and specificity of the model were 0.822 and 0.675, respectively, and the AUC was 0.829.Compared with the classic MELD score , the difference was not statistically significant (P>0.05).Conclusions The predictive model based on PCT and coagulation factors Ⅴand Ⅸcan effectively predict the short-term survival of cirrhotic patients complicated with SBP . The overall prognostic ability is not different from MELD score , but the model is more simple and easier to apply.
ABSTRACT
Objective To introduce the construction and application of clinical microbiology laboratory data management expert system.Methods Firstly, the process management was introduced to clinical microbiology laboratory. Then the characteristics of data on each node of work process were analyzed, and SQL Server data table was created as the knowledge base of the expert system.Finally, VB6.0 was used to compile the knowledge acquisition module, reasoning desktop module and input/output interface procedures to finally construct the expert system.Rates of defect report, errors in bacterial identification and drug sensitivity test, delay in culture results reporting and average delayed days were compared before and after the application of the expert system.Results The expert system could be used for data management in process nodes like sample reception, information collection and input, bacteria culture medium selection, bacterial identification and drug sensitive test, interpretation of drug sensitivity results, comprehensive evaluation in bacterial identification and drug sensitivity results, report of negative result, report of positive result, blood culture, Mycoplasma culture, time limit of detection, and nosocomial infection indicators.No defect report was found after the application of expert system; rate of errors in selection of drug sensitivity test medium was reduced from 0.81% ( 31/3 836 ) in 2012 to 0.02%(1/5 433) in 2014;rate of delay in culture results reporting was reduced from 1.78% (320/17 983) to 1.18%(232/19 692), and the average delayed days was also reduced (3.8 d vs.3.2 d).Conclusion Clinical microbiology laboratory data management expert system can improve work efficiency and reduce errors, which can enhance the overall management of laboratory and the quality of clinical service.
ABSTRACT
Objective To investigate the relationship between bacteria biofilm and bacterial culture in patients with chronic rhinosinusitis (CRS). Methods Ninety patients with CRS were enrolled in the study. Five patients with deviation of nasal septum and 10 healthy subjects served as controls. Mucosa of uncinate process or near the ostium of the maxillary sinus was obtained during endoscopic sinus surgery. All specimens were processed for bacterial culture and scanned by electron microscopy. Pearson test was performed to analyze the relationship between the presence of bacterial biofilm and the results of bacteria culture. Results The scanning electron microscopy showed bacterial biofilms in 64 (71.1%) out of 90patients with CRS, while the positive rate of bacteria culture in the study group was 66.7% (60/90). No bacterial biofilm and bacterium was detected in the control group and 26 culture-negative individuals in study group. Pearson correlation analysis showed a statistically association between bacterial biofilm and bacterial culture in CRS ( r = 0. 901, P = 0. 000). Conclusion Positive results of bacteria culture are highly correlated with the presence of bacterial biofilm in CRS patients.