RESUMO
OBJECTIVE: To compare clinical outcome before and after application of clinical practice guideline (CPG) for sepsis. METHOD: The study was an observational study with historical control. The data of pre-guideline period collected by retrospective review of medical records during August 1st to October 30th 2005 were compared to those of post-guideline period collected during February 1st to April 30th 2006. The primary outcome was in-hospital mortality and the secondary outcome were organ failure, ICU length of stay and hospital length of stay. Data analysis used the computer’s package for chi-square, and Fisher’s exact test with significant level at 0.05. RESULTS: One hundred and seventy three patients with sepsis or septic shock were included; 90 patients in pre-CPG period and 83 patients in post-CPG period. Baseline characteristics of patients in pre-CPG and post-CPG periods were comparable. The common causes of sepsis were pneumonia, urinary tract infection, gastrointestinal and intra-abdominal infection, consecutively. Hospital mortality rate was significantly reduced from 73.3% in pre-CPG period to 54.2% in post-CPG period (p=0.01). With regard to organ failure, acute respiratory distress syndrome was significantly increased from 26.7% to 44.9% (p=0.02) during the same period. Meanwhile, proportion of acute tubular necrosis and disseminated intravascular coagulopathy were found to be greater after introduction of the guideline. Furthermore, the length of ICU stay and total hospital stay trended longer in post-CPG period. CONCLUSIONS: The CPG for sepsis is effective in reducing in-hospital mortality. The reciprocal increase in morbidity, acute respiratory distress syndrome indicates that the CPG implementation might imperfect and requires further improvement.
RESUMO
PROBLEM/BACKGROUND: The avian influenza (H5N1) virus is an emerging virus that causes many human fatalities and poses an increasing pandemic threat. A surveillance system is an important part of the disease control of the pandemic influenza strategy. The initial screening guideline includes clinical manifestations, history of exposure and rapid test.OBJECTIVE: To determine the agreement of rapid test and RT-PCR for influenza in suspected cases of avian influenza, and final diagnosis of these cases.RESEARCH DESIGN: Diagnostic test study.SETTING: Budhacinnaraj Phitsanulok Hospital.MATERIALS and METHODS: A retrospective study of medical records of 137 consecutive patients presented to Budhachinnaraj Hospital screening scheme for avian influenza from Jan, 2005 to Dec, 2006. The rapid test was chosen to be the screening test. RT-PCR for influenza was performed as a confirmatory test. Statistical analysis was done using Kappa test, chi-square test, and Fisherís exact test; presented by sensitivity, specificity, positive predictive value and Kappa coefficient.RESULTS: The most common final diagnosis of suspected cases were non-specific URI (47.8%), followed by human influenza A (22.6%). With respect to clinical manifestations, history of fever and upper respiratory tract symptoms were the most common findings (92 % and 91.2 % respectively). The rapid test was performed in all suspected cases. RT-PCR for influenza was pursued in cases with high index of suspicious (72.3%), with 31.3% positive result. Sensitivity of rapid test compared to RT-PCR was unacceptable low (41.9%). Kappa coefficient was 0.476 (p \< 0.001).CONCLUSIONS: There is no avian influenza identified in our hospital during the study period. Human influenza were identified 22.6% of the suspected cases of avian influenza. Accuracy of rapid test from nasopharyngeal swab was found to be very low which raise awareness of interpretation of the test. The efficacy of the agreement of these two tests is medium.