Objective To study the risk factors of preoperative deep vein thrombosis(DVT)in middle-aged and elderly patients with lower limb fractures,and establish and evaluate the constructed nomogram model.Methods A retrospective analysis was conducted on the clinical data of totally 240 middle-aged and elderly patients with lower limb fractures at Quanzhou First Hospital Affiliated to Fujian Medical University from July 2023 to June 2024.According to whether the patient had DVT before surgery,they were divided into a DVT group of 75 cases and a non DVT group of 165 cases.The differences in clinical data and some labora-tory indicators levels at the time of initial admission,including platelets(PLT),prothrombin time(PT),ac-tive partial thromboplastin time(APTT),fibrinogen(FIB),thrombin time(TT),D-dimer(D-D),total pro-tein(TP),albumin(ALB),triglycerides(TG),cholesterol(CHOL),high-density lipoprotein cholesterol(HDL-C),low density lipoprotein cholesterol(LDL-C)between two groups were compared.Logistic binary regression was used to analyze the risk factors of DVT and a nomogram model was constructed.Receiver oper-ating characteristic(ROC)curve was used to evaluate the diagnostic efficacy of various risk factors and nomo-gram model in screening DVT.Results Compared age and D-D levels of the DVT group were significantly higher(P＜0.001),while the levels of APTT,TP and ALB were significantly lower(P＜0.05),and the pro-portion of DVT combined with internal diseases and the proportion of fractures in the femur were significantly higher(P＜0.001).The above-mentioned differential indicators were all single risk factors for DVT.ROC curve analysis showed that except for comorbidities of internal medicine,all other risk factors could effectively screen for DVT(P＜0.05),the area under the curve(AUC)was 0.573-0.706,and the constructed nomo-gram model could effectively improve AUC to 0.797.Conclusion The nomogram model constructed based on single risk factors for DVT in this study is helpful for clinical assessment of the risk of DVT occurrence in pa-tients at the early stage of admission,intuitively and effectively.

Objective To explore the predictive value of plasma cell-free DNA (cf-DNA) for prognosis in patients with sepsis. Methods 105 patients with sepsis admitted to department of emergency of the First Hospital of Quanzhou Affiliated to Fujian Medical University from June 2015 to June 2017 were enrolled. Patients were divided into sepsis group (n = 50) and severe sepsis group (n = 55). At the same time, 50 cases of physical examination center in our hospital were randomly selected as the healthy control group. The differences of cf-DNA, procalcitonin (PCT) and acute physiology and chronic health evaluation Ⅱ(APACHEⅡ) score among the three groups were compared. The correlation between cf-DNA and PCT or APACHEⅡ were analyzed by Bivarite method. Logistic regression was used to analyze the independent predictors of sepsis. The receiver operating characteristic curve (ROC) was made to evaluate cf-DNA, PCT, APACHEⅡ alone or combined ability to predict the prognosis of sepsis. Results The PCT, APACHE Ⅱ and cf-DNA in the sepsis group and severe sepsis group were significantly higher than those in the healthy control group [PCT (μg/L):5.80 (3.28, 8.85), 17.53 (8.40, 29.61) vs. 0.02 (0.01, 0.03); APACHEⅡ: 13.04±3.03, 23.29±8.02 vs. 2.10±1.05;cf-DNA (μg/L): 1 438.0 (1 154.0, 1 576.0), 2 595.0 (2 162.0, 5 198.0) vs. 17.0 (13.0, 20.5); all P﹤0.05], and the indicators in the severe sepsis group were further higher than the sepsis group (all P < 0.05). Correlation analysis showed that cf-DNA was significantly positive correlated with PCT [r = 0.675, 95% confidence interval (95%CI) = 0.575-0.766, P < 0.001] and APACHEⅡ (r = 0.911, 95%CI = 0.874-0.939, P < 0.001). ROC curve analysis showed that the areas under ROC curve (AUC) of PCT, APACHEⅡ, cf-DNA, PCT+APACHEⅡ, cf-DNA+PCT, cf-DNA+APACHEⅡ, cf-DNA+PCT+APACHEⅡ to predict the prognosis of sepsis patients were 0.898, 0.905, 0.961, 0.941, 0.974, 0.976 and 0.982, respectively. It was shown that when predicted alone with PCT, APACHEⅡ and cf-DNA, the AUC of cf-DNA was the largest (0.961), the sensitivity was 100%, and the specificity was 81.43%; the combined prediction of cf-DNA with PCT or APACHEⅡ could further increase AUC, and the combined prediction of cf-DNA, PCT and APACHEⅡhad the highest AUC (0.982), the sensitivity was 94.29%, the specificity was 98.57%. Conclusions cf-DNA, PCT and APACHEⅡ have certain predictive value for the prognosis of sepsis. The value of cf-DNA was the highest when predicted alone, but the predictive ability of cf-DNA combined with PCT and APACHEⅡ was the best.

Objective To develop a method of simultaneous detection and Gram classification for pathogens causing sepsis with gram-specific probes based real-time PCR. Methods A pair of universal primers and a set of probes including Gram-positive probe and Gram-negative probe were designed based on the bacterial highly conserved region of 16SrRNA gene. With the gram-specific probes based real-time PCR, 35 clinical frequently-isolated strains including 17 gram-positive and 18 gram-negative bacteria were identified correctly with the corresponding gram probe. The blood samples from 512 cases of suspected septicemia, who were hospitalized in our neonatal ward and the NICU and developed clinical signs suggestive of infection, were tested with routine culture and bacterial gram-specific probes based real-time PCR separately. Results The detection limit of the gram-specific probes based real-time PCR assay was 10 CFU of the bacteria. The 35 isolates could be detected and classified correctly by gram-specific probes based real-time PCR. The PCR results were all negative for Cytomegalo virus, EB virus, hepatitis B virus, Cryptococcns neoformans, candida albican, human genomic DNA and negative control. The gram-specific probes based real-time PCR appeared to be quite specific. For 512 blood specimens from the patients with suspicious neonatal sepsis, the positive rate of the gram-specific probes based real-time PCR array was 8.20% (42/512,), which is significantly higher than that of blood culture (32/512, 6.25%) (χ2=8.10, P<0.01). When blood culture was used as a standard, the sensitivity of the gram-specific probes based real-time PCR was 100%. The specificity was 97.92% and the accuracy was 98.05%. Canclusions Cram-specific probes based real-time PCR with universal primers and gram-specific probes are developed. This study suggests that the bacterial gram-specific probes based real-time PCR are very useful for the rapid and accurate diagnosis of bacterial infection.