Use of Procalcitonin to Help Diagnose the Patient Bacterial Infection in Sanpasitthiprasong Hospital, Ubon Ratchathani Province

Abstract

The objectives of this retrospective research were to study the results of the diagnosis of bacterial infection, the relationship between gender and age with the diagnostic results of bacterial infection, the specificity, the sensitivity, and the accuracy of procalcitonin diagnostic test of inpatients at Sunpasitthiprasong Hospital. The patients’ data were collected between July 2021 to December 2023. The sample group consisted of inpatients who requested for procalcitonin tests and had diagnostic results. A systematic random sampling was used to obtain a sample size of 480 persons. The research tool was a case record form. Data were analyzed for the general data of the inpatients and test specificity, test sensitivity, and area under the curve (AUC) using descriptive statistics including frequency, percentage, and median. The inferential statistics including 95% CI, Chi-Square, and Mann-Whitney U test were also used. The results of the study found female accounting for 51.67% and the median age was 60.50 years. Of 42.50% inpatients were medicine patients. The prevalence of the inpatients with bacterial infection was 70.83% (95% CI=68.73-72.93). According to the inferential statistic test, the diagnostic result of bacterial infection was not related to gender but was significantly related to age (p value=.036). When the procalcitonin cutoff point of 0.90 µg/L, the specificity and the sensitivity of procalcitonin diagnostic test were presented in percentage with 95%CI, 91.43% (95%CI=89.03-93.83) and 52.94% (95%CI=50.24-55.64) respectively. The good level of its accuracy discrimination was shown with AUC of 0.80. However, when the procalcitonin cut-off point was set at 0.50 µg/L, the test specificity was 82.86% (95%CI=79.06-86.66) and the test sensitivity was 63.53% (95%CI=60.93-66.13). Overall, this study confirms that the procalcitonin cut-off point at 0.90 µg/L had a high specificity of 91.43% with few false positives that might be used to differentiate between bacterial and viral infection with a high degree of accuracy. Therefore, the procalcitonin biomarker might be used for diagnose bacterial infections and to consider unnecessary antibiotic administration.