Accuracy of Procalcitonin Levels for Diagnosis of Culture-Positive Sepsis in Critically Ill Trauma Patients: A Retrospective Analysis.

Background Abdominal trauma and intra-abdominal sepsis are associated with significant morbidity and mortality. Microcirculation in the gut is disrupted in hemorrhagic and septic shock leading to tissue hypoxia, and the damaged gut acts as a reservoir rich in inflammatory mediators and provides a continual source of inflammation to the systemic circulation leading to sepsis. Sepsis is defined as the presence (probable or documented) of infection together with a systemic inflammatory response to infection. Blood culture is commonly considered to be the preferred approach for diagnosing sepsis, although it is time-consuming, that is, reports are normally available only after 12-48 hours. Procalcitonin levels (PCT) have recently emerged as a promising biomarker in the diagnosis of sepsis. The aim of the present study is to determine the diagnostic accuracy of PCT levels in predicting sepsis in critically ill trauma patients. Methodology This was designed as a validation study conducted in the Indoor Department of General Surgery, Liaquat National Hospital, Karachi. The sample size was calculated by taking the estimated frequency of sepsis in suspected patients at 62.13%, expected sensitivity of PCT at 70.83%, and specificity at 84.21% and the desired precision level of 12% for sensitivity; the calculated sample size was 96. The non-probability consecutive sampling method was used to recruit participants who were diagnosed with sepsis on clinical assessment. Blood culture samples were sent for the enrolled patients and a final diagnosis was made on the blood culture report. PCT levels were measured in these suspected patients on the same day of sending blood culture. Diagnostic accuracy of PCT size was measured using the receiver operating characteristic (ROC) curve. ROC curve was formulated for PCT levels against culture-proven sepsis to determine the ideal cut-off value of PCT levels. Two different cut-offs were determined to obtain the highest sensitivity and highest specificity accordingly. Results A total of 97 individuals met the inclusion criteria with a mean age of 34.89 ± 10.52 years. Mean PCT levels were 0.96 ± 0.59, with a gender predilection towards females (p < 0.001). No age difference was documented among gender (p = 0.655). The mean duration of intensive care unit stay was 11.73 ± 3.56 days. Culture-proven sepsis was identified in 67.0% of the study participants with a higher PCT level (p < 0.001). Among the 52.6% males included in the study, half were reported to have culture-positive sepsis, but among the 47.4% females culture was positive in 87% (p < 0.001). ROC revealed PCT was predictive for culture-positive sepsis at a cut-off value 0.47 ng/mL (p < 0.001), with a sensitivity of 92.3%, specificity of 68.7%, positive predictive value (PPV) of 85.7%, and negative predictive value (NPV) of 81.5%. By increasing the cut-off value to 0.90 ng/mL at area under the curve of 0.816, the specificity increased to 81.3% and sensitivity became 66.2%, with a PPV of 87.8% and NPV of 54.2%. Conclusion Our study determined two cut-values for PCT to predict sepsis, one with the highest sensitivity and the other with better specificity. Other than that, higher PCT levels were significant in female trauma patients. We conclude that PCT is a reliable marker for culture-proven diagnosis of sepsis and may aid physicians/surgeons to promptly manage patients accordingly.